[GH-ISSUE #14291] Model "Load failed" - "llama runner process has terminated: exit status 2" #9301

New Issue

GiteaMirror · 2026-04-12T22:10:15-05:00

GiteaMirror commented

2026-04-12 22:10:15 -05:00

Originally created by @TiiStyle on GitHub (Feb 17, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/14291

What is the issue?

Ollama fails to load model deepseek-r1:14b.
deepseek-r1:8b and deepseek-r1:32b do work flawlessly. -> RAM + VRAM sufficient for 14b.

LLama Server becomes unresponsive and crashes when trying to load the model.

Last 2 ollama versions before the one mentioned in version are also affected.
GPU: NVIDIA GeForce GTX 980 (is supported according to the docs)
CPU: Intel Core i7-6700K
RAM: 16 GB
VRAM: 4 GB
-> Allows running models up to 32B, when all other programs are closed. Gemma27B and QWEN30B also work fine.
But deepseek32B runs at 1 token per 68 seconds because Setup, therefore no option for usage, but it does load up and starts generating output.

-> I didn't have all programs closed on the test run, but that's also no different. Still same error. It's a model loading error for whatever reason.

Deleting and re-downloading the model didn't solve the issue.

Changing the context window size or running ollama through CMD using 'ollama serve' does not help.
No difference when using the ollama GUI versus API (generate + chat).

Need help, but better a fix. Will answer further questions happily.

Best regards

Relevant log output

time=2026-02-17T02:39:21.162+01:00 level=INFO source=routes.go:1663 msg="server config" env="map[CUDA_VISIBLE_DEVICES:0 GGML_VK_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:4096 OLLAMA_DEBUG:INFO OLLAMA_EDITOR: OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:F:\\OllamaModels OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NO_CLOUD:false OLLAMA_NUM_PARALLEL:1 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_REMOTES:[ollama.com] OLLAMA_SCHED_SPREAD:false OLLAMA_VULKAN:false ROCR_VISIBLE_DEVICES:]"
time=2026-02-17T02:39:21.381+01:00 level=INFO source=routes.go:1665 msg="Ollama cloud disabled: false"
time=2026-02-17T02:39:21.394+01:00 level=INFO source=images.go:473 msg="total blobs: 37"
time=2026-02-17T02:39:21.398+01:00 level=INFO source=images.go:480 msg="total unused blobs removed: 0"
time=2026-02-17T02:39:21.404+01:00 level=INFO source=routes.go:1718 msg="Listening on 127.0.0.1:11434 (version 0.16.2)"
time=2026-02-17T02:39:21.407+01:00 level=INFO source=runner.go:67 msg="discovering available GPUs..."
time=2026-02-17T02:39:21.465+01:00 level=WARN source=runner.go:485 msg="user overrode visible devices" CUDA_VISIBLE_DEVICES=0
time=2026-02-17T02:39:21.465+01:00 level=WARN source=runner.go:489 msg="if GPUs are not correctly discovered, unset and try again"
time=2026-02-17T02:39:21.483+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 26162"
time=2026-02-17T02:39:23.199+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 26186"
time=2026-02-17T02:39:24.592+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 26197"
time=2026-02-17T02:39:24.969+01:00 level=INFO source=runner.go:106 msg="experimental Vulkan support disabled.  To enable, set OLLAMA_VULKAN=1"
time=2026-02-17T02:39:24.971+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 26208"
time=2026-02-17T02:39:24.971+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 26209"
time=2026-02-17T02:39:25.295+01:00 level=INFO source=types.go:42 msg="inference compute" id=GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 filter_id="" library=CUDA compute=5.2 name=CUDA0 description="NVIDIA GeForce GTX 980" libdirs=ollama,cuda_v12 driver=13.0 pci_id=0000:01:00.0 type=discrete total="4.0 GiB" available="3.2 GiB"
time=2026-02-17T02:39:25.295+01:00 level=INFO source=routes.go:1768 msg="vram-based default context" total_vram="4.0 GiB" default_num_ctx=4096
[GIN] 2026/02/17 - 02:39:25 | 200 |       911.2µs |       127.0.0.1 | GET      "/api/version"
[GIN] 2026/02/17 - 02:39:25 | 200 |       911.2µs |       127.0.0.1 | GET      "/api/version"
[GIN] 2026/02/17 - 02:39:25 | 200 |            0s |       127.0.0.1 | GET      "/api/version"
[GIN] 2026/02/17 - 02:39:25 | 200 |      5.3327ms |       127.0.0.1 | GET      "/api/tags"
[GIN] 2026/02/17 - 02:39:25 | 200 |    136.0549ms |       127.0.0.1 | POST     "/api/show"
[GIN] 2026/02/17 - 02:39:25 | 401 |    192.3749ms |       127.0.0.1 | POST     "/api/me"
[GIN] 2026/02/17 - 02:39:25 | 401 |     194.202ms |       127.0.0.1 | POST     "/api/me"
[GIN] 2026/02/17 - 02:39:55 | 200 |      4.0126ms |       127.0.0.1 | GET      "/api/tags"
[GIN] 2026/02/17 - 02:39:59 | 200 |      5.3519ms |       127.0.0.1 | GET      "/api/tags"
[GIN] 2026/02/17 - 02:40:03 | 200 |      4.9056ms |       127.0.0.1 | GET      "/api/tags"
[GIN] 2026/02/17 - 02:40:07 | 200 |      5.2466ms |       127.0.0.1 | GET      "/api/tags"
[GIN] 2026/02/17 - 02:40:07 | 200 |    125.3406ms |       127.0.0.1 | POST     "/api/show"
[GIN] 2026/02/17 - 02:40:07 | 200 |    113.7989ms |       127.0.0.1 | POST     "/api/show"
time=2026-02-17T02:40:07.527+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 26266"
time=2026-02-17T02:40:07.808+01:00 level=INFO source=cpu_windows.go:148 msg=packages count=1
time=2026-02-17T02:40:07.808+01:00 level=INFO source=cpu_windows.go:195 msg="" package=0 cores=4 efficiency=0 threads=8
llama_model_loader: loaded meta data with 26 key-value pairs and 579 tensors from F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e (version GGUF V3 (latest))
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
llama_model_loader: - kv   0:                       general.architecture str              = qwen2
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                               general.name str              = DeepSeek R1 Distill Qwen 14B
llama_model_loader: - kv   3:                           general.basename str              = DeepSeek-R1-Distill-Qwen
llama_model_loader: - kv   4:                         general.size_label str              = 14B
llama_model_loader: - kv   5:                          qwen2.block_count u32              = 48
llama_model_loader: - kv   6:                       qwen2.context_length u32              = 131072
llama_model_loader: - kv   7:                     qwen2.embedding_length u32              = 5120
llama_model_loader: - kv   8:                  qwen2.feed_forward_length u32              = 13824
llama_model_loader: - kv   9:                 qwen2.attention.head_count u32              = 40
llama_model_loader: - kv  10:              qwen2.attention.head_count_kv u32              = 8
llama_model_loader: - kv  11:                       qwen2.rope.freq_base f32              = 1000000.000000
llama_model_loader: - kv  12:     qwen2.attention.layer_norm_rms_epsilon f32              = 0.000010
llama_model_loader: - kv  13:                          general.file_type u32              = 15
llama_model_loader: - kv  14:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  15:                         tokenizer.ggml.pre str              = qwen2
llama_model_loader: - kv  16:                      tokenizer.ggml.tokens arr[str,152064]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  17:                  tokenizer.ggml.token_type arr[i32,152064]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  18:                      tokenizer.ggml.merges arr[str,151387]  = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv  19:                tokenizer.ggml.bos_token_id u32              = 151646
llama_model_loader: - kv  20:                tokenizer.ggml.eos_token_id u32              = 151643
llama_model_loader: - kv  21:            tokenizer.ggml.padding_token_id u32              = 151643
llama_model_loader: - kv  22:               tokenizer.ggml.add_bos_token bool             = true
llama_model_loader: - kv  23:               tokenizer.ggml.add_eos_token bool             = false
llama_model_loader: - kv  24:                    tokenizer.chat_template str              = {% if not add_generation_prompt is de...
llama_model_loader: - kv  25:               general.quantization_version u32              = 2
llama_model_loader: - type  f32:  241 tensors
llama_model_loader: - type q4_K:  289 tensors
llama_model_loader: - type q6_K:   49 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q4_K - Medium
print_info: file size   = 8.37 GiB (4.87 BPW) 
load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
load: printing all EOG tokens:
load:   - 151643 ('<｜end▁of▁sentence｜>')
load:   - 151662 ('<|fim_pad|>')
load:   - 151663 ('<|repo_name|>')
load:   - 151664 ('<|file_sep|>')
load: special tokens cache size = 22
load: token to piece cache size = 0.9310 MB
print_info: arch             = qwen2
print_info: vocab_only       = 1
print_info: no_alloc         = 0
print_info: model type       = ?B
print_info: model params     = 14.77 B
print_info: general.name     = DeepSeek R1 Distill Qwen 14B
print_info: vocab type       = BPE
print_info: n_vocab          = 152064
print_info: n_merges         = 151387
print_info: BOS token        = 151646 '<｜begin▁of▁sentence｜>'
print_info: EOS token        = 151643 '<｜end▁of▁sentence｜>'
print_info: EOT token        = 151643 '<｜end▁of▁sentence｜>'
print_info: PAD token        = 151643 '<｜end▁of▁sentence｜>'
print_info: LF token         = 198 'Ċ'
print_info: FIM PRE token    = 151659 '<|fim_prefix|>'
print_info: FIM SUF token    = 151661 '<|fim_suffix|>'
print_info: FIM MID token    = 151660 '<|fim_middle|>'
print_info: FIM PAD token    = 151662 '<|fim_pad|>'
print_info: FIM REP token    = 151663 '<|repo_name|>'
print_info: FIM SEP token    = 151664 '<|file_sep|>'
print_info: EOG token        = 151643 '<｜end▁of▁sentence｜>'
print_info: EOG token        = 151662 '<|fim_pad|>'
print_info: EOG token        = 151663 '<|repo_name|>'
print_info: EOG token        = 151664 '<|file_sep|>'
print_info: max token length = 256
llama_model_load: vocab only - skipping tensors
time=2026-02-17T02:40:08.153+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --model F:\\OllamaModels\\blobs\\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e --port 26278"
time=2026-02-17T02:40:08.155+01:00 level=INFO source=sched.go:466 msg="system memory" total="15.9 GiB" free="7.9 GiB" free_swap="15.0 GiB"
time=2026-02-17T02:40:08.155+01:00 level=INFO source=sched.go:473 msg="gpu memory" id=GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 library=CUDA available="2.8 GiB" free="3.3 GiB" minimum="457.0 MiB" overhead="0 B"
time=2026-02-17T02:40:08.155+01:00 level=INFO source=server.go:498 msg="loading model" "model layers"=49 requested=-1
time=2026-02-17T02:40:08.156+01:00 level=INFO source=device.go:240 msg="model weights" device=CUDA0 size="1.4 GiB"
time=2026-02-17T02:40:08.156+01:00 level=INFO source=device.go:245 msg="model weights" device=CPU size="6.5 GiB"
time=2026-02-17T02:40:08.156+01:00 level=INFO source=device.go:251 msg="kv cache" device=CUDA0 size="144.0 MiB"
time=2026-02-17T02:40:08.156+01:00 level=INFO source=device.go:256 msg="kv cache" device=CPU size="624.0 MiB"
time=2026-02-17T02:40:08.156+01:00 level=INFO source=device.go:262 msg="compute graph" device=CUDA0 size="916.1 MiB"
time=2026-02-17T02:40:08.156+01:00 level=INFO source=device.go:272 msg="total memory" size="9.6 GiB"
time=2026-02-17T02:40:08.204+01:00 level=INFO source=runner.go:965 msg="starting go runner"
load_backend: loaded CPU backend from C:\Users\robob\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll
ggml_cuda_init: GGML_CUDA_FORCE_MMQ:    no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 1 CUDA devices:
  Device 0: NVIDIA GeForce GTX 980, compute capability 5.2, VMM: yes, ID: GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2
load_backend: loaded CUDA backend from C:\Users\robob\AppData\Local\Programs\Ollama\lib\ollama\cuda_v12\ggml-cuda.dll
time=2026-02-17T02:40:08.397+01:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 CUDA.0.ARCHS=500,520,600,610,700,750,800,860,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(clang)
time=2026-02-17T02:40:08.398+01:00 level=INFO source=runner.go:1001 msg="Server listening on 127.0.0.1:26278"
time=2026-02-17T02:40:08.403+01:00 level=INFO source=runner.go:895 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Auto KvSize:4096 KvCacheType: NumThreads:4 GPULayers:9[ID:GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 Layers:9(39..47)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2026-02-17T02:40:08.404+01:00 level=INFO source=server.go:1350 msg="waiting for llama runner to start responding"
time=2026-02-17T02:40:08.404+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server loading model"
ggml_backend_cuda_device_get_memory device GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 utilizing NVML memory reporting free: 3495686144 total: 4294967296
llama_model_load_from_file_impl: using device CUDA0 (NVIDIA GeForce GTX 980) (0000:01:00.0) - 3333 MiB free
llama_model_loader: loaded meta data with 26 key-value pairs and 579 tensors from F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e (version GGUF V3 (latest))
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
llama_model_loader: - kv   0:                       general.architecture str              = qwen2
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                               general.name str              = DeepSeek R1 Distill Qwen 14B
llama_model_loader: - kv   3:                           general.basename str              = DeepSeek-R1-Distill-Qwen
llama_model_loader: - kv   4:                         general.size_label str              = 14B
llama_model_loader: - kv   5:                          qwen2.block_count u32              = 48
llama_model_loader: - kv   6:                       qwen2.context_length u32              = 131072
llama_model_loader: - kv   7:                     qwen2.embedding_length u32              = 5120
llama_model_loader: - kv   8:                  qwen2.feed_forward_length u32              = 13824
llama_model_loader: - kv   9:                 qwen2.attention.head_count u32              = 40
llama_model_loader: - kv  10:              qwen2.attention.head_count_kv u32              = 8
llama_model_loader: - kv  11:                       qwen2.rope.freq_base f32              = 1000000.000000
llama_model_loader: - kv  12:     qwen2.attention.layer_norm_rms_epsilon f32              = 0.000010
llama_model_loader: - kv  13:                          general.file_type u32              = 15
llama_model_loader: - kv  14:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  15:                         tokenizer.ggml.pre str              = qwen2
llama_model_loader: - kv  16:                      tokenizer.ggml.tokens arr[str,152064]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  17:                  tokenizer.ggml.token_type arr[i32,152064]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  18:                      tokenizer.ggml.merges arr[str,151387]  = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv  19:                tokenizer.ggml.bos_token_id u32              = 151646
llama_model_loader: - kv  20:                tokenizer.ggml.eos_token_id u32              = 151643
llama_model_loader: - kv  21:            tokenizer.ggml.padding_token_id u32              = 151643
llama_model_loader: - kv  22:               tokenizer.ggml.add_bos_token bool             = true
llama_model_loader: - kv  23:               tokenizer.ggml.add_eos_token bool             = false
llama_model_loader: - kv  24:                    tokenizer.chat_template str              = {% if not add_generation_prompt is de...
llama_model_loader: - kv  25:               general.quantization_version u32              = 2
llama_model_loader: - type  f32:  241 tensors
llama_model_loader: - type q4_K:  289 tensors
llama_model_loader: - type q6_K:   49 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q4_K - Medium
print_info: file size   = 8.37 GiB (4.87 BPW) 
load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
load: printing all EOG tokens:
load:   - 151643 ('<｜end▁of▁sentence｜>')
load:   - 151662 ('<|fim_pad|>')
load:   - 151663 ('<|repo_name|>')
load:   - 151664 ('<|file_sep|>')
load: special tokens cache size = 22
load: token to piece cache size = 0.9310 MB
print_info: arch             = qwen2
print_info: vocab_only       = 0
print_info: no_alloc         = 0
print_info: n_ctx_train      = 131072
print_info: n_embd           = 5120
print_info: n_embd_inp       = 5120
print_info: n_layer          = 48
print_info: n_head           = 40
print_info: n_head_kv        = 8
print_info: n_rot            = 128
print_info: n_swa            = 0
print_info: is_swa_any       = 0
print_info: n_embd_head_k    = 128
print_info: n_embd_head_v    = 128
print_info: n_gqa            = 5
print_info: n_embd_k_gqa     = 1024
print_info: n_embd_v_gqa     = 1024
print_info: f_norm_eps       = 0.0e+00
print_info: f_norm_rms_eps   = 1.0e-05
print_info: f_clamp_kqv      = 0.0e+00
print_info: f_max_alibi_bias = 0.0e+00
print_info: f_logit_scale    = 0.0e+00
print_info: f_attn_scale     = 0.0e+00
print_info: n_ff             = 13824
print_info: n_expert         = 0
print_info: n_expert_used    = 0
print_info: n_expert_groups  = 0
print_info: n_group_used     = 0
print_info: causal attn      = 1
print_info: pooling type     = -1
print_info: rope type        = 2
print_info: rope scaling     = linear
print_info: freq_base_train  = 1000000.0
print_info: freq_scale_train = 1
print_info: n_ctx_orig_yarn  = 131072
print_info: rope_yarn_log_mul= 0.0000
print_info: rope_finetuned   = unknown
print_info: model type       = 14B
print_info: model params     = 14.77 B
print_info: general.name     = DeepSeek R1 Distill Qwen 14B
print_info: vocab type       = BPE
print_info: n_vocab          = 152064
print_info: n_merges         = 151387
print_info: BOS token        = 151646 '<｜begin▁of▁sentence｜>'
print_info: EOS token        = 151643 '<｜end▁of▁sentence｜>'
print_info: EOT token        = 151643 '<｜end▁of▁sentence｜>'
print_info: PAD token        = 151643 '<｜end▁of▁sentence｜>'
print_info: LF token         = 198 'Ċ'
print_info: FIM PRE token    = 151659 '<|fim_prefix|>'
print_info: FIM SUF token    = 151661 '<|fim_suffix|>'
print_info: FIM MID token    = 151660 '<|fim_middle|>'
print_info: FIM PAD token    = 151662 '<|fim_pad|>'
print_info: FIM REP token    = 151663 '<|repo_name|>'
print_info: FIM SEP token    = 151664 '<|file_sep|>'
print_info: EOG token        = 151643 '<｜end▁of▁sentence｜>'
print_info: EOG token        = 151662 '<|fim_pad|>'
print_info: EOG token        = 151663 '<|repo_name|>'
print_info: EOG token        = 151664 '<|file_sep|>'
print_info: max token length = 256
load_tensors: loading model tensors, this can take a while... (mmap = false)
load_tensors: offloading 9 repeating layers to GPU
load_tensors: offloaded 9/49 layers to GPU
load_tensors:          CPU model buffer size =   417.66 MiB
load_tensors:        CUDA0 model buffer size =  1460.34 MiB
load_tensors:    CUDA_Host model buffer size =  6688.04 MiB
llama_context: constructing llama_context
llama_context: n_seq_max     = 1
llama_context: n_ctx         = 4096
llama_context: n_ctx_seq     = 4096
llama_context: n_batch       = 512
llama_context: n_ubatch      = 512
llama_context: causal_attn   = 1
llama_context: flash_attn    = auto
llama_context: kv_unified    = false
llama_context: freq_base     = 1000000.0
llama_context: freq_scale    = 1
llama_context: n_ctx_seq (4096) < n_ctx_train (131072) -- the full capacity of the model will not be utilized
llama_context:        CPU  output buffer size =     0.60 MiB
llama_kv_cache:        CPU KV buffer size =   624.00 MiB
llama_kv_cache:      CUDA0 KV buffer size =   144.00 MiB
llama_kv_cache: size =  768.00 MiB (  4096 cells,  48 layers,  1/1 seqs), K (f16):  384.00 MiB, V (f16):  384.00 MiB
llama_context: Flash Attention was auto, set to enabled
graph_reserve: failed to allocate compute buffers
Exception 0xc0000005 0x0 0x1ce3b620d20 0x7ffce030e53e
PC=0x7ffce030e53e
signal arrived during external code execution

runtime.cgocall(0x7ff793e6e2b0, 0xc00035bc00)
	runtime/cgocall.go:167 +0x3e fp=0xc00035bbd8 sp=0xc00035bb70 pc=0x7ff792fd243e
github.com/ollama/ollama/llama._Cfunc_llama_init_from_model(0x1d3e93b74c0, {0x1000, 0x200, 0x200, 0x1, 0x4, 0x4, 0xffffffff, 0xffffffff, 0xffffffff, ...})
	_cgo_gotypes.go:759 +0x54 fp=0xc00035bc00 sp=0xc00035bbd8 pc=0x7ff793479774
github.com/ollama/ollama/llama.NewContextWithModel.func1(...)
	github.com/ollama/ollama/llama/llama.go:322
github.com/ollama/ollama/llama.NewContextWithModel(0xc00060ea60, {{0x1000, 0x200, 0x200, 0x1, 0x4, 0x4, 0xffffffff, 0xffffffff, 0xffffffff, ...}})
	github.com/ollama/ollama/llama/llama.go:322 +0x158 fp=0xc00035bda0 sp=0xc00035bc00 pc=0x7ff79347dd58
github.com/ollama/ollama/runner/llamarunner.(*Server).loadModel(0xc0004614a0, {{0xc00070a870, 0x1, 0x1}, 0x9, 0x0, 0x0, {0xc00070a868, 0x1, 0x2}, ...}, ...)
	github.com/ollama/ollama/runner/llamarunner/runner.go:847 +0x178 fp=0xc00035bee8 sp=0xc00035bda0 pc=0x7ff79352c938
github.com/ollama/ollama/runner/llamarunner.(*Server).load.gowrap2()
	github.com/ollama/ollama/runner/llamarunner/runner.go:934 +0x114 fp=0xc00035bfe0 sp=0xc00035bee8 pc=0x7ff79352db54
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc00035bfe8 sp=0xc00035bfe0 pc=0x7ff792fdd9a1
created by github.com/ollama/ollama/runner/llamarunner.(*Server).load in goroutine 9
	github.com/ollama/ollama/runner/llamarunner/runner.go:934 +0x889

goroutine 1 gp=0xc0000021c0 m=nil [IO wait]:
runtime.gopark(0x7ff792fdf1a0?, 0x7ff795180d00?, 0x20?, 0xd9?, 0xc00046d9cc?)
	runtime/proc.go:435 +0xce fp=0xc00047d630 sp=0xc00047d610 pc=0x7ff792fd598e
runtime.netpollblock(0x3b8?, 0x92f70406?, 0xf7?)
	runtime/netpoll.go:575 +0xf7 fp=0xc00047d668 sp=0xc00047d630 pc=0x7ff792f9bdf7
internal/poll.runtime_pollWait(0x1d3e68fa8f0, 0x72)
	runtime/netpoll.go:351 +0x85 fp=0xc00047d688 sp=0xc00047d668 pc=0x7ff792fd4b25
internal/poll.(*pollDesc).wait(0x7ff79306a8f3?, 0x0?, 0x0)
	internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc00047d6b0 sp=0xc00047d688 pc=0x7ff79306bee7
internal/poll.execIO(0xc00046d920, 0xc00047d758)
	internal/poll/fd_windows.go:177 +0x105 fp=0xc00047d728 sp=0xc00047d6b0 pc=0x7ff79306d345
internal/poll.(*FD).acceptOne(0xc00046d908, 0x3c4, {0xc0002320f0?, 0xc00047d7b8?, 0x7ff793075005?}, 0xc00047d7ec?)
	internal/poll/fd_windows.go:946 +0x65 fp=0xc00047d788 sp=0xc00047d728 pc=0x7ff7930718c5
internal/poll.(*FD).Accept(0xc00046d908, 0xc00047d938)
	internal/poll/fd_windows.go:980 +0x1b6 fp=0xc00047d840 sp=0xc00047d788 pc=0x7ff793071bf6
net.(*netFD).accept(0xc00046d908)
	net/fd_windows.go:182 +0x4b fp=0xc00047d958 sp=0xc00047d840 pc=0x7ff7930e344b
net.(*TCPListener).accept(0xc0004b38c0)
	net/tcpsock_posix.go:159 +0x1b fp=0xc00047d9a8 sp=0xc00047d958 pc=0x7ff7930f99fb
net.(*TCPListener).Accept(0xc0004b38c0)
	net/tcpsock.go:380 +0x30 fp=0xc00047d9d8 sp=0xc00047d9a8 pc=0x7ff7930f87b0
net/http.(*onceCloseListener).Accept(0xc000204480?)
	<autogenerated>:1 +0x24 fp=0xc00047d9f0 sp=0xc00047d9d8 pc=0x7ff793311ec4
net/http.(*Server).Serve(0xc0006af700, {0x7ff7946e21d0, 0xc0004b38c0})
	net/http/server.go:3424 +0x30c fp=0xc00047db20 sp=0xc00047d9f0 pc=0x7ff7932e978c
github.com/ollama/ollama/runner/llamarunner.Execute({0xc0000a2020, 0x4, 0x6})
	github.com/ollama/ollama/runner/llamarunner/runner.go:1002 +0x8f5 fp=0xc00047dcf0 sp=0xc00047db20 pc=0x7ff79352e515
github.com/ollama/ollama/runner.Execute({0xc0000a2010?, 0x0?, 0x0?})
	github.com/ollama/ollama/runner/runner.go:25 +0x1a5 fp=0xc00047dd30 sp=0xc00047dcf0 pc=0x7ff793603ae5
github.com/ollama/ollama/cmd.NewCLI.func3(0xc0006af200?, {0x7ff7944c4587?, 0x4?, 0x7ff7944c458b?})
	github.com/ollama/ollama/cmd/cmd.go:2262 +0x45 fp=0xc00047dd58 sp=0xc00047dd30 pc=0x7ff793dfce65
github.com/spf13/cobra.(*Command).execute(0xc000147508, {0xc0004b3640, 0x4, 0x4})
	github.com/spf13/cobra@v1.7.0/command.go:940 +0x85c fp=0xc00047de78 sp=0xc00047dd58 pc=0x7ff79315e61c
github.com/spf13/cobra.(*Command).ExecuteC(0xc0006ee908)
	github.com/spf13/cobra@v1.7.0/command.go:1068 +0x3a5 fp=0xc00047df30 sp=0xc00047de78 pc=0x7ff79315ee65
github.com/spf13/cobra.(*Command).Execute(...)
	github.com/spf13/cobra@v1.7.0/command.go:992
github.com/spf13/cobra.(*Command).ExecuteContext(...)
	github.com/spf13/cobra@v1.7.0/command.go:985
main.main()
	github.com/ollama/ollama/main.go:12 +0x4d fp=0xc00047df50 sp=0xc00047df30 pc=0x7ff793dff28d
runtime.main()
	runtime/proc.go:283 +0x27d fp=0xc00047dfe0 sp=0xc00047df50 pc=0x7ff792fa4ddd
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc00047dfe8 sp=0xc00047dfe0 pc=0x7ff792fdd9a1

goroutine 2 gp=0xc0000028c0 m=nil [force gc (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
	runtime/proc.go:435 +0xce fp=0xc00006ffa8 sp=0xc00006ff88 pc=0x7ff792fd598e
runtime.goparkunlock(...)
	runtime/proc.go:441
runtime.forcegchelper()
	runtime/proc.go:348 +0xb8 fp=0xc00006ffe0 sp=0xc00006ffa8 pc=0x7ff792fa50f8
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc00006ffe8 sp=0xc00006ffe0 pc=0x7ff792fdd9a1
created by runtime.init.7 in goroutine 1
	runtime/proc.go:336 +0x1a

goroutine 3 gp=0xc000002c40 m=nil [GC sweep wait]:
runtime.gopark(0x1?, 0x0?, 0x0?, 0x0?, 0x0?)
	runtime/proc.go:435 +0xce fp=0xc000071f80 sp=0xc000071f60 pc=0x7ff792fd598e
runtime.goparkunlock(...)
	runtime/proc.go:441
runtime.bgsweep(0xc00007e000)
	runtime/mgcsweep.go:316 +0xdf fp=0xc000071fc8 sp=0xc000071f80 pc=0x7ff792f8debf
runtime.gcenable.gowrap1()
	runtime/mgc.go:204 +0x25 fp=0xc000071fe0 sp=0xc000071fc8 pc=0x7ff792f82285
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc000071fe8 sp=0xc000071fe0 pc=0x7ff792fdd9a1
created by runtime.gcenable in goroutine 1
	runtime/mgc.go:204 +0x66

goroutine 4 gp=0xc000002e00 m=nil [GC scavenge wait]:
runtime.gopark(0x10000?, 0x7ff7946ccc38?, 0x0?, 0x0?, 0x0?)
	runtime/proc.go:435 +0xce fp=0xc000085f78 sp=0xc000085f58 pc=0x7ff792fd598e
runtime.goparkunlock(...)
	runtime/proc.go:441
runtime.(*scavengerState).park(0x7ff7951aa900)
	runtime/mgcscavenge.go:425 +0x49 fp=0xc000085fa8 sp=0xc000085f78 pc=0x7ff792f8b909
runtime.bgscavenge(0xc00007e000)
	runtime/mgcscavenge.go:658 +0x59 fp=0xc000085fc8 sp=0xc000085fa8 pc=0x7ff792f8be99
runtime.gcenable.gowrap2()
	runtime/mgc.go:205 +0x25 fp=0xc000085fe0 sp=0xc000085fc8 pc=0x7ff792f82225
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc000085fe8 sp=0xc000085fe0 pc=0x7ff792fdd9a1
created by runtime.gcenable in goroutine 1
	runtime/mgc.go:205 +0xa5

goroutine 5 gp=0xc000003340 m=nil [finalizer wait]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
	runtime/proc.go:435 +0xce fp=0xc000087e30 sp=0xc000087e10 pc=0x7ff792fd598e
runtime.runfinq()
	runtime/mfinal.go:196 +0x107 fp=0xc000087fe0 sp=0xc000087e30 pc=0x7ff792f81207
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc000087fe8 sp=0xc000087fe0 pc=0x7ff792fdd9a1
created by runtime.createfing in goroutine 1
	runtime/mfinal.go:166 +0x3d

goroutine 6 gp=0xc000003dc0 m=nil [chan receive]:
runtime.gopark(0xc00017b2c0?, 0xc000508018?, 0x60?, 0x3f?, 0x7ff7930cc148?)
	runtime/proc.go:435 +0xce fp=0xc000073f18 sp=0xc000073ef8 pc=0x7ff792fd598e
runtime.chanrecv(0xc00008a380, 0x0, 0x1)
	runtime/chan.go:664 +0x445 fp=0xc000073f90 sp=0xc000073f18 pc=0x7ff792f72d45
runtime.chanrecv1(0x7ff792fa4f40?, 0xc000073f76?)
	runtime/chan.go:506 +0x12 fp=0xc000073fb8 sp=0xc000073f90 pc=0x7ff792f728d2
runtime.unique_runtime_registerUniqueMapCleanup.func2(...)
	runtime/mgc.go:1796
runtime.unique_runtime_registerUniqueMapCleanup.gowrap1()
	runtime/mgc.go:1799 +0x2f fp=0xc000073fe0 sp=0xc000073fb8 pc=0x7ff792f854af
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc000073fe8 sp=0xc000073fe0 pc=0x7ff792fdd9a1
created by unique.runtime_registerUniqueMapCleanup in goroutine 1
	runtime/mgc.go:1794 +0x85

goroutine 7 gp=0xc0003d8380 m=nil [GC worker (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
	runtime/proc.go:435 +0xce fp=0xc000081f38 sp=0xc000081f18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008b960)
	runtime/mgc.go:1423 +0xe9 fp=0xc000081fc8 sp=0xc000081f38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
	runtime/mgc.go:1339 +0x25 fp=0xc000081fe0 sp=0xc000081fc8 pc=0x7ff792f84685
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc000081fe8 sp=0xc000081fe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
	runtime/mgc.go:1339 +0x105

goroutine 18 gp=0xc0004861c0 m=nil [GC worker (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
	runtime/proc.go:435 +0xce fp=0xc000491f38 sp=0xc000491f18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008b960)
	runtime/mgc.go:1423 +0xe9 fp=0xc000491fc8 sp=0xc000491f38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
	runtime/mgc.go:1339 +0x25 fp=0xc000491fe0 sp=0xc000491fc8 pc=0x7ff792f84685
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc000491fe8 sp=0xc000491fe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
	runtime/mgc.go:1339 +0x105

goroutine 19 gp=0xc000486380 m=nil [GC worker (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
	runtime/proc.go:435 +0xce fp=0xc000493f38 sp=0xc000493f18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008b960)
	runtime/mgc.go:1423 +0xe9 fp=0xc000493fc8 sp=0xc000493f38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
	runtime/mgc.go:1339 +0x25 fp=0xc000493fe0 sp=0xc000493fc8 pc=0x7ff792f84685
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc000493fe8 sp=0xc000493fe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
	runtime/mgc.go:1339 +0x105

goroutine 20 gp=0xc000486540 m=nil [GC worker (idle)]:
runtime.gopark(0x1f0ee5732bbc?, 0x0?, 0x0?, 0x0?, 0x0?)
	runtime/proc.go:435 +0xce fp=0xc00048df38 sp=0xc00048df18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008b960)
	runtime/mgc.go:1423 +0xe9 fp=0xc00048dfc8 sp=0xc00048df38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
	runtime/mgc.go:1339 +0x25 fp=0xc00048dfe0 sp=0xc00048dfc8 pc=0x7ff792f84685
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc00048dfe8 sp=0xc00048dfe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
	runtime/mgc.go:1339 +0x105

goroutine 34 gp=0xc000206000 m=nil [GC worker (idle)]:
runtime.gopark(0x7ff7951fc600?, 0x3?, 0x10?, 0x10?, 0x0?)
	runtime/proc.go:435 +0xce fp=0xc00020df38 sp=0xc00020df18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008b960)
	runtime/mgc.go:1423 +0xe9 fp=0xc00020dfc8 sp=0xc00020df38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
	runtime/mgc.go:1339 +0x25 fp=0xc00020dfe0 sp=0xc00020dfc8 pc=0x7ff792f84685
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc00020dfe8 sp=0xc00020dfe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
	runtime/mgc.go:1339 +0x105

goroutine 35 gp=0xc0002061c0 m=nil [GC worker (idle)]:
runtime.gopark(0x1f0ee5732bbc?, 0x0?, 0x0?, 0x0?, 0x0?)
	runtime/proc.go:435 +0xce fp=0xc00020ff38 sp=0xc00020ff18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008b960)
	runtime/mgc.go:1423 +0xe9 fp=0xc00020ffc8 sp=0xc00020ff38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
	runtime/mgc.go:1339 +0x25 fp=0xc00020ffe0 sp=0xc00020ffc8 pc=0x7ff792f84685
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc00020ffe8 sp=0xc00020ffe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
	runtime/mgc.go:1339 +0x105

goroutine 36 gp=0xc000206380 m=nil [GC worker (idle)]:
runtime.gopark(0x1f0ee5732bbc?, 0x0?, 0x0?, 0x0?, 0x0?)
	runtime/proc.go:435 +0xce fp=0xc000209f38 sp=0xc000209f18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008b960)
	runtime/mgc.go:1423 +0xe9 fp=0xc000209fc8 sp=0xc000209f38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
	runtime/mgc.go:1339 +0x25 fp=0xc000209fe0 sp=0xc000209fc8 pc=0x7ff792f84685
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc000209fe8 sp=0xc000209fe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
	runtime/mgc.go:1339 +0x105

goroutine 21 gp=0xc000486700 m=nil [GC worker (idle)]:
runtime.gopark(0x1f0ee5732bbc?, 0x0?, 0x0?, 0x0?, 0x0?)
	runtime/proc.go:435 +0xce fp=0xc00048ff38 sp=0xc00048ff18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008b960)
	runtime/mgc.go:1423 +0xe9 fp=0xc00048ffc8 sp=0xc00048ff38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
	runtime/mgc.go:1339 +0x25 fp=0xc00048ffe0 sp=0xc00048ffc8 pc=0x7ff792f84685
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc00048ffe8 sp=0xc00048ffe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
	runtime/mgc.go:1339 +0x105

goroutine 8 gp=0xc000586700 m=nil [sync.WaitGroup.Wait]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x80?, 0x0?)
	runtime/proc.go:435 +0xce fp=0xc00020be20 sp=0xc00020be00 pc=0x7ff792fd598e
runtime.goparkunlock(...)
	runtime/proc.go:441
runtime.semacquire1(0xc0004614c0, 0x0, 0x1, 0x0, 0x18)
	runtime/sema.go:188 +0x22f fp=0xc00020be88 sp=0xc00020be20 pc=0x7ff792fb750f
sync.runtime_SemacquireWaitGroup(0x0?)
	runtime/sema.go:110 +0x25 fp=0xc00020bec0 sp=0xc00020be88 pc=0x7ff792fd6f85
sync.(*WaitGroup).Wait(0x0?)
	sync/waitgroup.go:118 +0x48 fp=0xc00020bee8 sp=0xc00020bec0 pc=0x7ff792feb988
github.com/ollama/ollama/runner/llamarunner.(*Server).run(0xc0004614a0, {0x7ff7946e4a30, 0xc0004a71d0})
	github.com/ollama/ollama/runner/llamarunner/runner.go:360 +0x4b fp=0xc00020bfb8 sp=0xc00020bee8 pc=0x7ff79352922b
github.com/ollama/ollama/runner/llamarunner.Execute.gowrap1()
	github.com/ollama/ollama/runner/llamarunner/runner.go:981 +0x28 fp=0xc00020bfe0 sp=0xc00020bfb8 pc=0x7ff79352e788
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc00020bfe8 sp=0xc00020bfe0 pc=0x7ff792fdd9a1
created by github.com/ollama/ollama/runner/llamarunner.Execute in goroutine 1
	github.com/ollama/ollama/runner/llamarunner/runner.go:981 +0x4c5

goroutine 9 gp=0xc0005868c0 m=nil [IO wait]:
runtime.gopark(0x0?, 0xc00046dba0?, 0x48?, 0xdc?, 0xc00046dc4c?)
	runtime/proc.go:435 +0xce fp=0xc0000498c8 sp=0xc0000498a8 pc=0x7ff792fd598e
runtime.netpollblock(0x3c0?, 0x92f70406?, 0xf7?)
	runtime/netpoll.go:575 +0xf7 fp=0xc000049900 sp=0xc0000498c8 pc=0x7ff792f9bdf7
internal/poll.runtime_pollWait(0x1d3e68fa7d8, 0x72)
	runtime/netpoll.go:351 +0x85 fp=0xc000049920 sp=0xc000049900 pc=0x7ff792fd4b25
internal/poll.(*pollDesc).wait(0x7ff79319d5f7?, 0xc000049970?, 0x0)
	internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc000049948 sp=0xc000049920 pc=0x7ff79306bee7
internal/poll.execIO(0xc00046dba0, 0x7ff794546098)
	internal/poll/fd_windows.go:177 +0x105 fp=0xc0000499c0 sp=0xc000049948 pc=0x7ff79306d345
internal/poll.(*FD).Read(0xc00046db88, {0xc00008f000, 0x1000, 0x1000})
	internal/poll/fd_windows.go:438 +0x29b fp=0xc000049a60 sp=0xc0000499c0 pc=0x7ff79306e01b
net.(*netFD).Read(0xc00046db88, {0xc00008f000?, 0xc000049ad0?, 0x7ff79306c3a5?})
	net/fd_posix.go:55 +0x25 fp=0xc000049aa8 sp=0xc000049a60 pc=0x7ff7930e1325
net.(*conn).Read(0xc0004a0350, {0xc00008f000?, 0x0?, 0x0?})
	net/net.go:194 +0x45 fp=0xc000049af0 sp=0xc000049aa8 pc=0x7ff7930f0a45
net/http.(*connReader).Read(0xc0000d39b0, {0xc00008f000, 0x1000, 0x1000})
	net/http/server.go:798 +0x159 fp=0xc000049b40 sp=0xc000049af0 pc=0x7ff7932de639
bufio.(*Reader).fill(0xc000193e60)
	bufio/bufio.go:113 +0x103 fp=0xc000049b78 sp=0xc000049b40 pc=0x7ff793107963
bufio.(*Reader).Peek(0xc000193e60, 0x4)
	bufio/bufio.go:152 +0x53 fp=0xc000049b98 sp=0xc000049b78 pc=0x7ff793107a93
net/http.(*conn).serve(0xc000204480, {0x7ff7946e49f8, 0xc000709020})
	net/http/server.go:2137 +0x785 fp=0xc000049fb8 sp=0xc000049b98 pc=0x7ff7932e4425
net/http.(*Server).Serve.gowrap3()
	net/http/server.go:3454 +0x28 fp=0xc000049fe0 sp=0xc000049fb8 pc=0x7ff7932e9b88
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc000049fe8 sp=0xc000049fe0 pc=0x7ff792fdd9a1
created by net/http.(*Server).Serve in goroutine 1
	net/http/server.go:3454 +0x485
rax     0xfffffffa521db500
rbx     0x1d3e9445820
rcx     0x1d3e93d3860
rdx     0x0
rdi     0x0
rsi     0x1d3e944b470
rbp     0x94aacffce0
rsp     0x94aacffa30
r8      0x1d3a11a0e40
r9      0x1
r10     0x8000
r11     0x94aacffa40
r12     0x667
r13     0x1218
r14     0x1
r15     0x1d3e93b74c0
rip     0x7ffce030e53e
rflags  0x10287
cs      0x33
fs      0x53
gs      0x2b
time=2026-02-17T02:40:31.914+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server not responding"
time=2026-02-17T02:40:32.165+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server error"
time=2026-02-17T02:40:32.666+01:00 level=INFO source=sched.go:493 msg="Load failed" model=F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e error="llama runner process has terminated: exit status 2"

OS

Windows

GPU

Nvidia

CPU

Intel

Ollama version

0.16.2

Originally created by @TiiStyle on GitHub (Feb 17, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/14291 ### What is the issue? Ollama fails to load model deepseek-r1:14b. deepseek-r1:8b and deepseek-r1:32b do work flawlessly. -> RAM + VRAM sufficient for 14b. LLama Server becomes unresponsive and crashes when trying to load the model. Last 2 ollama versions before the one mentioned in version are also affected. GPU: NVIDIA GeForce GTX 980 (is supported according to the docs) CPU: Intel Core i7-6700K RAM: 16 GB VRAM: 4 GB -> Allows running models up to 32B, when all other programs are closed. Gemma27B and QWEN30B also work fine. But deepseek32B runs at 1 token per 68 seconds because Setup, therefore no option for usage, but it does load up and starts generating output. -> I didn't have all programs closed on the test run, but that's also no different. Still same error. It's a model loading error for whatever reason. Deleting and re-downloading the model didn't solve the issue. Changing the context window size or running ollama through CMD using 'ollama serve' does not help. No difference when using the ollama GUI versus API (generate + chat). Need help, but better a fix. Will answer further questions happily. Best regards ### Relevant log output ```shell time=2026-02-17T02:39:21.162+01:00 level=INFO source=routes.go:1663 msg="server config" env="map[CUDA_VISIBLE_DEVICES:0 GGML_VK_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:4096 OLLAMA_DEBUG:INFO OLLAMA_EDITOR: OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:F:\\OllamaModels OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NO_CLOUD:false OLLAMA_NUM_PARALLEL:1 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_REMOTES:[ollama.com] OLLAMA_SCHED_SPREAD:false OLLAMA_VULKAN:false ROCR_VISIBLE_DEVICES:]" time=2026-02-17T02:39:21.381+01:00 level=INFO source=routes.go:1665 msg="Ollama cloud disabled: false" time=2026-02-17T02:39:21.394+01:00 level=INFO source=images.go:473 msg="total blobs: 37" time=2026-02-17T02:39:21.398+01:00 level=INFO source=images.go:480 msg="total unused blobs removed: 0" time=2026-02-17T02:39:21.404+01:00 level=INFO source=routes.go:1718 msg="Listening on 127.0.0.1:11434 (version 0.16.2)" time=2026-02-17T02:39:21.407+01:00 level=INFO source=runner.go:67 msg="discovering available GPUs..." time=2026-02-17T02:39:21.465+01:00 level=WARN source=runner.go:485 msg="user overrode visible devices" CUDA_VISIBLE_DEVICES=0 time=2026-02-17T02:39:21.465+01:00 level=WARN source=runner.go:489 msg="if GPUs are not correctly discovered, unset and try again" time=2026-02-17T02:39:21.483+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 26162" time=2026-02-17T02:39:23.199+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 26186" time=2026-02-17T02:39:24.592+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 26197" time=2026-02-17T02:39:24.969+01:00 level=INFO source=runner.go:106 msg="experimental Vulkan support disabled. To enable, set OLLAMA_VULKAN=1" time=2026-02-17T02:39:24.971+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 26208" time=2026-02-17T02:39:24.971+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 26209" time=2026-02-17T02:39:25.295+01:00 level=INFO source=types.go:42 msg="inference compute" id=GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 filter_id="" library=CUDA compute=5.2 name=CUDA0 description="NVIDIA GeForce GTX 980" libdirs=ollama,cuda_v12 driver=13.0 pci_id=0000:01:00.0 type=discrete total="4.0 GiB" available="3.2 GiB" time=2026-02-17T02:39:25.295+01:00 level=INFO source=routes.go:1768 msg="vram-based default context" total_vram="4.0 GiB" default_num_ctx=4096 [GIN] 2026/02/17 - 02:39:25 | 200 | 911.2µs | 127.0.0.1 | GET "/api/version" [GIN] 2026/02/17 - 02:39:25 | 200 | 911.2µs | 127.0.0.1 | GET "/api/version" [GIN] 2026/02/17 - 02:39:25 | 200 | 0s | 127.0.0.1 | GET "/api/version" [GIN] 2026/02/17 - 02:39:25 | 200 | 5.3327ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 02:39:25 | 200 | 136.0549ms | 127.0.0.1 | POST "/api/show" [GIN] 2026/02/17 - 02:39:25 | 401 | 192.3749ms | 127.0.0.1 | POST "/api/me" [GIN] 2026/02/17 - 02:39:25 | 401 | 194.202ms | 127.0.0.1 | POST "/api/me" [GIN] 2026/02/17 - 02:39:55 | 200 | 4.0126ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 02:39:59 | 200 | 5.3519ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 02:40:03 | 200 | 4.9056ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 02:40:07 | 200 | 5.2466ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 02:40:07 | 200 | 125.3406ms | 127.0.0.1 | POST "/api/show" [GIN] 2026/02/17 - 02:40:07 | 200 | 113.7989ms | 127.0.0.1 | POST "/api/show" time=2026-02-17T02:40:07.527+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 26266" time=2026-02-17T02:40:07.808+01:00 level=INFO source=cpu_windows.go:148 msg=packages count=1 time=2026-02-17T02:40:07.808+01:00 level=INFO source=cpu_windows.go:195 msg="" package=0 cores=4 efficiency=0 threads=8 llama_model_loader: loaded meta data with 26 key-value pairs and 579 tensors from F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e (version GGUF V3 (latest)) llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output. llama_model_loader: - kv 0: general.architecture str = qwen2 llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.name str = DeepSeek R1 Distill Qwen 14B llama_model_loader: - kv 3: general.basename str = DeepSeek-R1-Distill-Qwen llama_model_loader: - kv 4: general.size_label str = 14B llama_model_loader: - kv 5: qwen2.block_count u32 = 48 llama_model_loader: - kv 6: qwen2.context_length u32 = 131072 llama_model_loader: - kv 7: qwen2.embedding_length u32 = 5120 llama_model_loader: - kv 8: qwen2.feed_forward_length u32 = 13824 llama_model_loader: - kv 9: qwen2.attention.head_count u32 = 40 llama_model_loader: - kv 10: qwen2.attention.head_count_kv u32 = 8 llama_model_loader: - kv 11: qwen2.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 12: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 13: general.file_type u32 = 15 llama_model_loader: - kv 14: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 15: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 16: tokenizer.ggml.tokens arr[str,152064] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 18: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 151646 llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151643 llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 24: tokenizer.chat_template str = {% if not add_generation_prompt is de... llama_model_loader: - kv 25: general.quantization_version u32 = 2 llama_model_loader: - type f32: 241 tensors llama_model_loader: - type q4_K: 289 tensors llama_model_loader: - type q6_K: 49 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 8.37 GiB (4.87 BPW) load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect load: printing all EOG tokens: load: - 151643 ('<｜end▁of▁sentence｜>') load: - 151662 ('<|fim_pad|>') load: - 151663 ('<|repo_name|>') load: - 151664 ('<|file_sep|>') load: special tokens cache size = 22 load: token to piece cache size = 0.9310 MB print_info: arch = qwen2 print_info: vocab_only = 1 print_info: no_alloc = 0 print_info: model type = ?B print_info: model params = 14.77 B print_info: general.name = DeepSeek R1 Distill Qwen 14B print_info: vocab type = BPE print_info: n_vocab = 152064 print_info: n_merges = 151387 print_info: BOS token = 151646 '<｜begin▁of▁sentence｜>' print_info: EOS token = 151643 '<｜end▁of▁sentence｜>' print_info: EOT token = 151643 '<｜end▁of▁sentence｜>' print_info: PAD token = 151643 '<｜end▁of▁sentence｜>' print_info: LF token = 198 'Ċ' print_info: FIM PRE token = 151659 '<|fim_prefix|>' print_info: FIM SUF token = 151661 '<|fim_suffix|>' print_info: FIM MID token = 151660 '<|fim_middle|>' print_info: FIM PAD token = 151662 '<|fim_pad|>' print_info: FIM REP token = 151663 '<|repo_name|>' print_info: FIM SEP token = 151664 '<|file_sep|>' print_info: EOG token = 151643 '<｜end▁of▁sentence｜>' print_info: EOG token = 151662 '<|fim_pad|>' print_info: EOG token = 151663 '<|repo_name|>' print_info: EOG token = 151664 '<|file_sep|>' print_info: max token length = 256 llama_model_load: vocab only - skipping tensors time=2026-02-17T02:40:08.153+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --model F:\\OllamaModels\\blobs\\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e --port 26278" time=2026-02-17T02:40:08.155+01:00 level=INFO source=sched.go:466 msg="system memory" total="15.9 GiB" free="7.9 GiB" free_swap="15.0 GiB" time=2026-02-17T02:40:08.155+01:00 level=INFO source=sched.go:473 msg="gpu memory" id=GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 library=CUDA available="2.8 GiB" free="3.3 GiB" minimum="457.0 MiB" overhead="0 B" time=2026-02-17T02:40:08.155+01:00 level=INFO source=server.go:498 msg="loading model" "model layers"=49 requested=-1 time=2026-02-17T02:40:08.156+01:00 level=INFO source=device.go:240 msg="model weights" device=CUDA0 size="1.4 GiB" time=2026-02-17T02:40:08.156+01:00 level=INFO source=device.go:245 msg="model weights" device=CPU size="6.5 GiB" time=2026-02-17T02:40:08.156+01:00 level=INFO source=device.go:251 msg="kv cache" device=CUDA0 size="144.0 MiB" time=2026-02-17T02:40:08.156+01:00 level=INFO source=device.go:256 msg="kv cache" device=CPU size="624.0 MiB" time=2026-02-17T02:40:08.156+01:00 level=INFO source=device.go:262 msg="compute graph" device=CUDA0 size="916.1 MiB" time=2026-02-17T02:40:08.156+01:00 level=INFO source=device.go:272 msg="total memory" size="9.6 GiB" time=2026-02-17T02:40:08.204+01:00 level=INFO source=runner.go:965 msg="starting go runner" load_backend: loaded CPU backend from C:\Users\robob\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no ggml_cuda_init: found 1 CUDA devices: Device 0: NVIDIA GeForce GTX 980, compute capability 5.2, VMM: yes, ID: GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 load_backend: loaded CUDA backend from C:\Users\robob\AppData\Local\Programs\Ollama\lib\ollama\cuda_v12\ggml-cuda.dll time=2026-02-17T02:40:08.397+01:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 CUDA.0.ARCHS=500,520,600,610,700,750,800,860,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(clang) time=2026-02-17T02:40:08.398+01:00 level=INFO source=runner.go:1001 msg="Server listening on 127.0.0.1:26278" time=2026-02-17T02:40:08.403+01:00 level=INFO source=runner.go:895 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Auto KvSize:4096 KvCacheType: NumThreads:4 GPULayers:9[ID:GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 Layers:9(39..47)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2026-02-17T02:40:08.404+01:00 level=INFO source=server.go:1350 msg="waiting for llama runner to start responding" time=2026-02-17T02:40:08.404+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server loading model" ggml_backend_cuda_device_get_memory device GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 utilizing NVML memory reporting free: 3495686144 total: 4294967296 llama_model_load_from_file_impl: using device CUDA0 (NVIDIA GeForce GTX 980) (0000:01:00.0) - 3333 MiB free llama_model_loader: loaded meta data with 26 key-value pairs and 579 tensors from F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e (version GGUF V3 (latest)) llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output. llama_model_loader: - kv 0: general.architecture str = qwen2 llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.name str = DeepSeek R1 Distill Qwen 14B llama_model_loader: - kv 3: general.basename str = DeepSeek-R1-Distill-Qwen llama_model_loader: - kv 4: general.size_label str = 14B llama_model_loader: - kv 5: qwen2.block_count u32 = 48 llama_model_loader: - kv 6: qwen2.context_length u32 = 131072 llama_model_loader: - kv 7: qwen2.embedding_length u32 = 5120 llama_model_loader: - kv 8: qwen2.feed_forward_length u32 = 13824 llama_model_loader: - kv 9: qwen2.attention.head_count u32 = 40 llama_model_loader: - kv 10: qwen2.attention.head_count_kv u32 = 8 llama_model_loader: - kv 11: qwen2.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 12: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 13: general.file_type u32 = 15 llama_model_loader: - kv 14: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 15: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 16: tokenizer.ggml.tokens arr[str,152064] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 18: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 151646 llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151643 llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 24: tokenizer.chat_template str = {% if not add_generation_prompt is de... llama_model_loader: - kv 25: general.quantization_version u32 = 2 llama_model_loader: - type f32: 241 tensors llama_model_loader: - type q4_K: 289 tensors llama_model_loader: - type q6_K: 49 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 8.37 GiB (4.87 BPW) load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect load: printing all EOG tokens: load: - 151643 ('<｜end▁of▁sentence｜>') load: - 151662 ('<|fim_pad|>') load: - 151663 ('<|repo_name|>') load: - 151664 ('<|file_sep|>') load: special tokens cache size = 22 load: token to piece cache size = 0.9310 MB print_info: arch = qwen2 print_info: vocab_only = 0 print_info: no_alloc = 0 print_info: n_ctx_train = 131072 print_info: n_embd = 5120 print_info: n_embd_inp = 5120 print_info: n_layer = 48 print_info: n_head = 40 print_info: n_head_kv = 8 print_info: n_rot = 128 print_info: n_swa = 0 print_info: is_swa_any = 0 print_info: n_embd_head_k = 128 print_info: n_embd_head_v = 128 print_info: n_gqa = 5 print_info: n_embd_k_gqa = 1024 print_info: n_embd_v_gqa = 1024 print_info: f_norm_eps = 0.0e+00 print_info: f_norm_rms_eps = 1.0e-05 print_info: f_clamp_kqv = 0.0e+00 print_info: f_max_alibi_bias = 0.0e+00 print_info: f_logit_scale = 0.0e+00 print_info: f_attn_scale = 0.0e+00 print_info: n_ff = 13824 print_info: n_expert = 0 print_info: n_expert_used = 0 print_info: n_expert_groups = 0 print_info: n_group_used = 0 print_info: causal attn = 1 print_info: pooling type = -1 print_info: rope type = 2 print_info: rope scaling = linear print_info: freq_base_train = 1000000.0 print_info: freq_scale_train = 1 print_info: n_ctx_orig_yarn = 131072 print_info: rope_yarn_log_mul= 0.0000 print_info: rope_finetuned = unknown print_info: model type = 14B print_info: model params = 14.77 B print_info: general.name = DeepSeek R1 Distill Qwen 14B print_info: vocab type = BPE print_info: n_vocab = 152064 print_info: n_merges = 151387 print_info: BOS token = 151646 '<｜begin▁of▁sentence｜>' print_info: EOS token = 151643 '<｜end▁of▁sentence｜>' print_info: EOT token = 151643 '<｜end▁of▁sentence｜>' print_info: PAD token = 151643 '<｜end▁of▁sentence｜>' print_info: LF token = 198 'Ċ' print_info: FIM PRE token = 151659 '<|fim_prefix|>' print_info: FIM SUF token = 151661 '<|fim_suffix|>' print_info: FIM MID token = 151660 '<|fim_middle|>' print_info: FIM PAD token = 151662 '<|fim_pad|>' print_info: FIM REP token = 151663 '<|repo_name|>' print_info: FIM SEP token = 151664 '<|file_sep|>' print_info: EOG token = 151643 '<｜end▁of▁sentence｜>' print_info: EOG token = 151662 '<|fim_pad|>' print_info: EOG token = 151663 '<|repo_name|>' print_info: EOG token = 151664 '<|file_sep|>' print_info: max token length = 256 load_tensors: loading model tensors, this can take a while... (mmap = false) load_tensors: offloading 9 repeating layers to GPU load_tensors: offloaded 9/49 layers to GPU load_tensors: CPU model buffer size = 417.66 MiB load_tensors: CUDA0 model buffer size = 1460.34 MiB load_tensors: CUDA_Host model buffer size = 6688.04 MiB llama_context: constructing llama_context llama_context: n_seq_max = 1 llama_context: n_ctx = 4096 llama_context: n_ctx_seq = 4096 llama_context: n_batch = 512 llama_context: n_ubatch = 512 llama_context: causal_attn = 1 llama_context: flash_attn = auto llama_context: kv_unified = false llama_context: freq_base = 1000000.0 llama_context: freq_scale = 1 llama_context: n_ctx_seq (4096) < n_ctx_train (131072) -- the full capacity of the model will not be utilized llama_context: CPU output buffer size = 0.60 MiB llama_kv_cache: CPU KV buffer size = 624.00 MiB llama_kv_cache: CUDA0 KV buffer size = 144.00 MiB llama_kv_cache: size = 768.00 MiB ( 4096 cells, 48 layers, 1/1 seqs), K (f16): 384.00 MiB, V (f16): 384.00 MiB llama_context: Flash Attention was auto, set to enabled graph_reserve: failed to allocate compute buffers Exception 0xc0000005 0x0 0x1ce3b620d20 0x7ffce030e53e PC=0x7ffce030e53e signal arrived during external code execution runtime.cgocall(0x7ff793e6e2b0, 0xc00035bc00) runtime/cgocall.go:167 +0x3e fp=0xc00035bbd8 sp=0xc00035bb70 pc=0x7ff792fd243e github.com/ollama/ollama/llama._Cfunc_llama_init_from_model(0x1d3e93b74c0, {0x1000, 0x200, 0x200, 0x1, 0x4, 0x4, 0xffffffff, 0xffffffff, 0xffffffff, ...}) _cgo_gotypes.go:759 +0x54 fp=0xc00035bc00 sp=0xc00035bbd8 pc=0x7ff793479774 github.com/ollama/ollama/llama.NewContextWithModel.func1(...) github.com/ollama/ollama/llama/llama.go:322 github.com/ollama/ollama/llama.NewContextWithModel(0xc00060ea60, {{0x1000, 0x200, 0x200, 0x1, 0x4, 0x4, 0xffffffff, 0xffffffff, 0xffffffff, ...}}) github.com/ollama/ollama/llama/llama.go:322 +0x158 fp=0xc00035bda0 sp=0xc00035bc00 pc=0x7ff79347dd58 github.com/ollama/ollama/runner/llamarunner.(*Server).loadModel(0xc0004614a0, {{0xc00070a870, 0x1, 0x1}, 0x9, 0x0, 0x0, {0xc00070a868, 0x1, 0x2}, ...}, ...) github.com/ollama/ollama/runner/llamarunner/runner.go:847 +0x178 fp=0xc00035bee8 sp=0xc00035bda0 pc=0x7ff79352c938 github.com/ollama/ollama/runner/llamarunner.(*Server).load.gowrap2() github.com/ollama/ollama/runner/llamarunner/runner.go:934 +0x114 fp=0xc00035bfe0 sp=0xc00035bee8 pc=0x7ff79352db54 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc00035bfe8 sp=0xc00035bfe0 pc=0x7ff792fdd9a1 created by github.com/ollama/ollama/runner/llamarunner.(*Server).load in goroutine 9 github.com/ollama/ollama/runner/llamarunner/runner.go:934 +0x889 goroutine 1 gp=0xc0000021c0 m=nil [IO wait]: runtime.gopark(0x7ff792fdf1a0?, 0x7ff795180d00?, 0x20?, 0xd9?, 0xc00046d9cc?) runtime/proc.go:435 +0xce fp=0xc00047d630 sp=0xc00047d610 pc=0x7ff792fd598e runtime.netpollblock(0x3b8?, 0x92f70406?, 0xf7?) runtime/netpoll.go:575 +0xf7 fp=0xc00047d668 sp=0xc00047d630 pc=0x7ff792f9bdf7 internal/poll.runtime_pollWait(0x1d3e68fa8f0, 0x72) runtime/netpoll.go:351 +0x85 fp=0xc00047d688 sp=0xc00047d668 pc=0x7ff792fd4b25 internal/poll.(*pollDesc).wait(0x7ff79306a8f3?, 0x0?, 0x0) internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc00047d6b0 sp=0xc00047d688 pc=0x7ff79306bee7 internal/poll.execIO(0xc00046d920, 0xc00047d758) internal/poll/fd_windows.go:177 +0x105 fp=0xc00047d728 sp=0xc00047d6b0 pc=0x7ff79306d345 internal/poll.(*FD).acceptOne(0xc00046d908, 0x3c4, {0xc0002320f0?, 0xc00047d7b8?, 0x7ff793075005?}, 0xc00047d7ec?) internal/poll/fd_windows.go:946 +0x65 fp=0xc00047d788 sp=0xc00047d728 pc=0x7ff7930718c5 internal/poll.(*FD).Accept(0xc00046d908, 0xc00047d938) internal/poll/fd_windows.go:980 +0x1b6 fp=0xc00047d840 sp=0xc00047d788 pc=0x7ff793071bf6 net.(*netFD).accept(0xc00046d908) net/fd_windows.go:182 +0x4b fp=0xc00047d958 sp=0xc00047d840 pc=0x7ff7930e344b net.(*TCPListener).accept(0xc0004b38c0) net/tcpsock_posix.go:159 +0x1b fp=0xc00047d9a8 sp=0xc00047d958 pc=0x7ff7930f99fb net.(*TCPListener).Accept(0xc0004b38c0) net/tcpsock.go:380 +0x30 fp=0xc00047d9d8 sp=0xc00047d9a8 pc=0x7ff7930f87b0 net/http.(*onceCloseListener).Accept(0xc000204480?) <autogenerated>:1 +0x24 fp=0xc00047d9f0 sp=0xc00047d9d8 pc=0x7ff793311ec4 net/http.(*Server).Serve(0xc0006af700, {0x7ff7946e21d0, 0xc0004b38c0}) net/http/server.go:3424 +0x30c fp=0xc00047db20 sp=0xc00047d9f0 pc=0x7ff7932e978c github.com/ollama/ollama/runner/llamarunner.Execute({0xc0000a2020, 0x4, 0x6}) github.com/ollama/ollama/runner/llamarunner/runner.go:1002 +0x8f5 fp=0xc00047dcf0 sp=0xc00047db20 pc=0x7ff79352e515 github.com/ollama/ollama/runner.Execute({0xc0000a2010?, 0x0?, 0x0?}) github.com/ollama/ollama/runner/runner.go:25 +0x1a5 fp=0xc00047dd30 sp=0xc00047dcf0 pc=0x7ff793603ae5 github.com/ollama/ollama/cmd.NewCLI.func3(0xc0006af200?, {0x7ff7944c4587?, 0x4?, 0x7ff7944c458b?}) github.com/ollama/ollama/cmd/cmd.go:2262 +0x45 fp=0xc00047dd58 sp=0xc00047dd30 pc=0x7ff793dfce65 github.com/spf13/cobra.(*Command).execute(0xc000147508, {0xc0004b3640, 0x4, 0x4}) github.com/spf13/cobra@v1.7.0/command.go:940 +0x85c fp=0xc00047de78 sp=0xc00047dd58 pc=0x7ff79315e61c github.com/spf13/cobra.(*Command).ExecuteC(0xc0006ee908) github.com/spf13/cobra@v1.7.0/command.go:1068 +0x3a5 fp=0xc00047df30 sp=0xc00047de78 pc=0x7ff79315ee65 github.com/spf13/cobra.(*Command).Execute(...) github.com/spf13/cobra@v1.7.0/command.go:992 github.com/spf13/cobra.(*Command).ExecuteContext(...) github.com/spf13/cobra@v1.7.0/command.go:985 main.main() github.com/ollama/ollama/main.go:12 +0x4d fp=0xc00047df50 sp=0xc00047df30 pc=0x7ff793dff28d runtime.main() runtime/proc.go:283 +0x27d fp=0xc00047dfe0 sp=0xc00047df50 pc=0x7ff792fa4ddd runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc00047dfe8 sp=0xc00047dfe0 pc=0x7ff792fdd9a1 goroutine 2 gp=0xc0000028c0 m=nil [force gc (idle)]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc00006ffa8 sp=0xc00006ff88 pc=0x7ff792fd598e runtime.goparkunlock(...) runtime/proc.go:441 runtime.forcegchelper() runtime/proc.go:348 +0xb8 fp=0xc00006ffe0 sp=0xc00006ffa8 pc=0x7ff792fa50f8 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc00006ffe8 sp=0xc00006ffe0 pc=0x7ff792fdd9a1 created by runtime.init.7 in goroutine 1 runtime/proc.go:336 +0x1a goroutine 3 gp=0xc000002c40 m=nil [GC sweep wait]: runtime.gopark(0x1?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000071f80 sp=0xc000071f60 pc=0x7ff792fd598e runtime.goparkunlock(...) runtime/proc.go:441 runtime.bgsweep(0xc00007e000) runtime/mgcsweep.go:316 +0xdf fp=0xc000071fc8 sp=0xc000071f80 pc=0x7ff792f8debf runtime.gcenable.gowrap1() runtime/mgc.go:204 +0x25 fp=0xc000071fe0 sp=0xc000071fc8 pc=0x7ff792f82285 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000071fe8 sp=0xc000071fe0 pc=0x7ff792fdd9a1 created by runtime.gcenable in goroutine 1 runtime/mgc.go:204 +0x66 goroutine 4 gp=0xc000002e00 m=nil [GC scavenge wait]: runtime.gopark(0x10000?, 0x7ff7946ccc38?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000085f78 sp=0xc000085f58 pc=0x7ff792fd598e runtime.goparkunlock(...) runtime/proc.go:441 runtime.(*scavengerState).park(0x7ff7951aa900) runtime/mgcscavenge.go:425 +0x49 fp=0xc000085fa8 sp=0xc000085f78 pc=0x7ff792f8b909 runtime.bgscavenge(0xc00007e000) runtime/mgcscavenge.go:658 +0x59 fp=0xc000085fc8 sp=0xc000085fa8 pc=0x7ff792f8be99 runtime.gcenable.gowrap2() runtime/mgc.go:205 +0x25 fp=0xc000085fe0 sp=0xc000085fc8 pc=0x7ff792f82225 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000085fe8 sp=0xc000085fe0 pc=0x7ff792fdd9a1 created by runtime.gcenable in goroutine 1 runtime/mgc.go:205 +0xa5 goroutine 5 gp=0xc000003340 m=nil [finalizer wait]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000087e30 sp=0xc000087e10 pc=0x7ff792fd598e runtime.runfinq() runtime/mfinal.go:196 +0x107 fp=0xc000087fe0 sp=0xc000087e30 pc=0x7ff792f81207 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000087fe8 sp=0xc000087fe0 pc=0x7ff792fdd9a1 created by runtime.createfing in goroutine 1 runtime/mfinal.go:166 +0x3d goroutine 6 gp=0xc000003dc0 m=nil [chan receive]: runtime.gopark(0xc00017b2c0?, 0xc000508018?, 0x60?, 0x3f?, 0x7ff7930cc148?) runtime/proc.go:435 +0xce fp=0xc000073f18 sp=0xc000073ef8 pc=0x7ff792fd598e runtime.chanrecv(0xc00008a380, 0x0, 0x1) runtime/chan.go:664 +0x445 fp=0xc000073f90 sp=0xc000073f18 pc=0x7ff792f72d45 runtime.chanrecv1(0x7ff792fa4f40?, 0xc000073f76?) runtime/chan.go:506 +0x12 fp=0xc000073fb8 sp=0xc000073f90 pc=0x7ff792f728d2 runtime.unique_runtime_registerUniqueMapCleanup.func2(...) runtime/mgc.go:1796 runtime.unique_runtime_registerUniqueMapCleanup.gowrap1() runtime/mgc.go:1799 +0x2f fp=0xc000073fe0 sp=0xc000073fb8 pc=0x7ff792f854af runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000073fe8 sp=0xc000073fe0 pc=0x7ff792fdd9a1 created by unique.runtime_registerUniqueMapCleanup in goroutine 1 runtime/mgc.go:1794 +0x85 goroutine 7 gp=0xc0003d8380 m=nil [GC worker (idle)]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000081f38 sp=0xc000081f18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008b960) runtime/mgc.go:1423 +0xe9 fp=0xc000081fc8 sp=0xc000081f38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc000081fe0 sp=0xc000081fc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000081fe8 sp=0xc000081fe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 18 gp=0xc0004861c0 m=nil [GC worker (idle)]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000491f38 sp=0xc000491f18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008b960) runtime/mgc.go:1423 +0xe9 fp=0xc000491fc8 sp=0xc000491f38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc000491fe0 sp=0xc000491fc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000491fe8 sp=0xc000491fe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 19 gp=0xc000486380 m=nil [GC worker (idle)]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000493f38 sp=0xc000493f18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008b960) runtime/mgc.go:1423 +0xe9 fp=0xc000493fc8 sp=0xc000493f38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc000493fe0 sp=0xc000493fc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000493fe8 sp=0xc000493fe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 20 gp=0xc000486540 m=nil [GC worker (idle)]: runtime.gopark(0x1f0ee5732bbc?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc00048df38 sp=0xc00048df18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008b960) runtime/mgc.go:1423 +0xe9 fp=0xc00048dfc8 sp=0xc00048df38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc00048dfe0 sp=0xc00048dfc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc00048dfe8 sp=0xc00048dfe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 34 gp=0xc000206000 m=nil [GC worker (idle)]: runtime.gopark(0x7ff7951fc600?, 0x3?, 0x10?, 0x10?, 0x0?) runtime/proc.go:435 +0xce fp=0xc00020df38 sp=0xc00020df18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008b960) runtime/mgc.go:1423 +0xe9 fp=0xc00020dfc8 sp=0xc00020df38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc00020dfe0 sp=0xc00020dfc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc00020dfe8 sp=0xc00020dfe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 35 gp=0xc0002061c0 m=nil [GC worker (idle)]: runtime.gopark(0x1f0ee5732bbc?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc00020ff38 sp=0xc00020ff18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008b960) runtime/mgc.go:1423 +0xe9 fp=0xc00020ffc8 sp=0xc00020ff38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc00020ffe0 sp=0xc00020ffc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc00020ffe8 sp=0xc00020ffe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 36 gp=0xc000206380 m=nil [GC worker (idle)]: runtime.gopark(0x1f0ee5732bbc?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000209f38 sp=0xc000209f18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008b960) runtime/mgc.go:1423 +0xe9 fp=0xc000209fc8 sp=0xc000209f38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc000209fe0 sp=0xc000209fc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000209fe8 sp=0xc000209fe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 21 gp=0xc000486700 m=nil [GC worker (idle)]: runtime.gopark(0x1f0ee5732bbc?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc00048ff38 sp=0xc00048ff18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008b960) runtime/mgc.go:1423 +0xe9 fp=0xc00048ffc8 sp=0xc00048ff38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc00048ffe0 sp=0xc00048ffc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc00048ffe8 sp=0xc00048ffe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 8 gp=0xc000586700 m=nil [sync.WaitGroup.Wait]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x80?, 0x0?) runtime/proc.go:435 +0xce fp=0xc00020be20 sp=0xc00020be00 pc=0x7ff792fd598e runtime.goparkunlock(...) runtime/proc.go:441 runtime.semacquire1(0xc0004614c0, 0x0, 0x1, 0x0, 0x18) runtime/sema.go:188 +0x22f fp=0xc00020be88 sp=0xc00020be20 pc=0x7ff792fb750f sync.runtime_SemacquireWaitGroup(0x0?) runtime/sema.go:110 +0x25 fp=0xc00020bec0 sp=0xc00020be88 pc=0x7ff792fd6f85 sync.(*WaitGroup).Wait(0x0?) sync/waitgroup.go:118 +0x48 fp=0xc00020bee8 sp=0xc00020bec0 pc=0x7ff792feb988 github.com/ollama/ollama/runner/llamarunner.(*Server).run(0xc0004614a0, {0x7ff7946e4a30, 0xc0004a71d0}) github.com/ollama/ollama/runner/llamarunner/runner.go:360 +0x4b fp=0xc00020bfb8 sp=0xc00020bee8 pc=0x7ff79352922b github.com/ollama/ollama/runner/llamarunner.Execute.gowrap1() github.com/ollama/ollama/runner/llamarunner/runner.go:981 +0x28 fp=0xc00020bfe0 sp=0xc00020bfb8 pc=0x7ff79352e788 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc00020bfe8 sp=0xc00020bfe0 pc=0x7ff792fdd9a1 created by github.com/ollama/ollama/runner/llamarunner.Execute in goroutine 1 github.com/ollama/ollama/runner/llamarunner/runner.go:981 +0x4c5 goroutine 9 gp=0xc0005868c0 m=nil [IO wait]: runtime.gopark(0x0?, 0xc00046dba0?, 0x48?, 0xdc?, 0xc00046dc4c?) runtime/proc.go:435 +0xce fp=0xc0000498c8 sp=0xc0000498a8 pc=0x7ff792fd598e runtime.netpollblock(0x3c0?, 0x92f70406?, 0xf7?) runtime/netpoll.go:575 +0xf7 fp=0xc000049900 sp=0xc0000498c8 pc=0x7ff792f9bdf7 internal/poll.runtime_pollWait(0x1d3e68fa7d8, 0x72) runtime/netpoll.go:351 +0x85 fp=0xc000049920 sp=0xc000049900 pc=0x7ff792fd4b25 internal/poll.(*pollDesc).wait(0x7ff79319d5f7?, 0xc000049970?, 0x0) internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc000049948 sp=0xc000049920 pc=0x7ff79306bee7 internal/poll.execIO(0xc00046dba0, 0x7ff794546098) internal/poll/fd_windows.go:177 +0x105 fp=0xc0000499c0 sp=0xc000049948 pc=0x7ff79306d345 internal/poll.(*FD).Read(0xc00046db88, {0xc00008f000, 0x1000, 0x1000}) internal/poll/fd_windows.go:438 +0x29b fp=0xc000049a60 sp=0xc0000499c0 pc=0x7ff79306e01b net.(*netFD).Read(0xc00046db88, {0xc00008f000?, 0xc000049ad0?, 0x7ff79306c3a5?}) net/fd_posix.go:55 +0x25 fp=0xc000049aa8 sp=0xc000049a60 pc=0x7ff7930e1325 net.(*conn).Read(0xc0004a0350, {0xc00008f000?, 0x0?, 0x0?}) net/net.go:194 +0x45 fp=0xc000049af0 sp=0xc000049aa8 pc=0x7ff7930f0a45 net/http.(*connReader).Read(0xc0000d39b0, {0xc00008f000, 0x1000, 0x1000}) net/http/server.go:798 +0x159 fp=0xc000049b40 sp=0xc000049af0 pc=0x7ff7932de639 bufio.(*Reader).fill(0xc000193e60) bufio/bufio.go:113 +0x103 fp=0xc000049b78 sp=0xc000049b40 pc=0x7ff793107963 bufio.(*Reader).Peek(0xc000193e60, 0x4) bufio/bufio.go:152 +0x53 fp=0xc000049b98 sp=0xc000049b78 pc=0x7ff793107a93 net/http.(*conn).serve(0xc000204480, {0x7ff7946e49f8, 0xc000709020}) net/http/server.go:2137 +0x785 fp=0xc000049fb8 sp=0xc000049b98 pc=0x7ff7932e4425 net/http.(*Server).Serve.gowrap3() net/http/server.go:3454 +0x28 fp=0xc000049fe0 sp=0xc000049fb8 pc=0x7ff7932e9b88 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000049fe8 sp=0xc000049fe0 pc=0x7ff792fdd9a1 created by net/http.(*Server).Serve in goroutine 1 net/http/server.go:3454 +0x485 rax 0xfffffffa521db500 rbx 0x1d3e9445820 rcx 0x1d3e93d3860 rdx 0x0 rdi 0x0 rsi 0x1d3e944b470 rbp 0x94aacffce0 rsp 0x94aacffa30 r8 0x1d3a11a0e40 r9 0x1 r10 0x8000 r11 0x94aacffa40 r12 0x667 r13 0x1218 r14 0x1 r15 0x1d3e93b74c0 rip 0x7ffce030e53e rflags 0x10287 cs 0x33 fs 0x53 gs 0x2b time=2026-02-17T02:40:31.914+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server not responding" time=2026-02-17T02:40:32.165+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server error" time=2026-02-17T02:40:32.666+01:00 level=INFO source=sched.go:493 msg="Load failed" model=F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e error="llama runner process has terminated: exit status 2" ``` ### OS Windows ### GPU Nvidia ### CPU Intel ### Ollama version 0.16.2

GiteaMirror added the bug label 2026-04-12 22:10:15 -05:00

GiteaMirror commented

2026-04-12 22:10:16 -05:00

@rick-github commented on GitHub (Feb 17, 2026):

load_tensors: offloaded 9/49 layers to GPU
graph_reserve: failed to allocate compute buffers

The runner thought it had enough VRAM to load 9 layers but seems like that was too much. See here for ways to mitigate OOM issues.

@rick-github commented on GitHub (Feb 17, 2026): ``` load_tensors: offloaded 9/49 layers to GPU graph_reserve: failed to allocate compute buffers ``` The runner thought it had enough VRAM to load 9 layers but seems like that was too much. See [here](https://github.com/ollama/ollama/issues/8597#issuecomment-2614533288) for ways to mitigate OOM issues.

GiteaMirror commented

2026-04-12 22:10:17 -05:00

@TiiStyle commented on GitHub (Feb 17, 2026):

The runner thought it had enough VRAM to load 9 layers but seems like that was too much. See https://github.com/ollama/ollama/issues/8597#issuecomment-2614533288 for ways to mitigate OOM issues.

The issue still persists as long as the GPU is used at all.
I created multiple modelfiles, where I used num_gpu to load less layers onto the GPU. ALL of them failed!!! Even loading just one single layer to the GPU fails.
Only when setting num_gpu to ZERO, then the model loads and starts generating output. In all cases with num_gpu > 0, it fails relentlessly.

And since I haven't encountered any other model where this behaviour occurs, I still am confident that there is some kind of actual problem underlying the specific setup I use, mostly because it loads deepseek-r1:8b and deepseek-r1:32b without any issues and without the creation of modelfiles.

@TiiStyle commented on GitHub (Feb 17, 2026): > The runner thought it had enough VRAM to load 9 layers but seems like that was too much. See https://github.com/ollama/ollama/issues/8597#issuecomment-2614533288 for ways to mitigate OOM issues. The issue still persists as long as the GPU is used at all. I created multiple modelfiles, where I used num_gpu to load less layers onto the GPU. ALL of them failed!!! Even loading just one single layer to the GPU fails. Only when setting num_gpu to ZERO, then the model loads and starts generating output. In all cases with num_gpu > 0, it fails relentlessly. And since I haven't encountered any other model where this behaviour occurs, I still am confident that there is some kind of actual problem underlying the specific setup I use, mostly because it loads deepseek-r1:8b and deepseek-r1:32b without any issues and without the creation of modelfiles.

GiteaMirror commented

2026-04-12 22:10:17 -05:00

@rick-github commented on GitHub (Feb 17, 2026):

Logs?

@rick-github commented on GitHub (Feb 17, 2026): Logs?

GiteaMirror commented

2026-04-12 22:10:18 -05:00

@TiiStyle commented on GitHub (Feb 17, 2026):

Here are the combined logs for:
num_gpu: 4, 2, 1 and 0.

time=2026-02-17T16:09:47.967+01:00 level=INFO source=routes.go:1663 msg="server config" env="map[CUDA_VISIBLE_DEVICES:0 GGML_VK_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:4096 OLLAMA_DEBUG:INFO OLLAMA_EDITOR: OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:F:\OllamaModels OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NO_CLOUD:false OLLAMA_NUM_PARALLEL:1 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_REMOTES:[ollama.com] OLLAMA_SCHED_SPREAD:false OLLAMA_VULKAN:false ROCR_VISIBLE_DEVICES:]"
time=2026-02-17T16:09:48.077+01:00 level=INFO source=routes.go:1665 msg="Ollama cloud disabled: false"
time=2026-02-17T16:09:48.095+01:00 level=INFO source=images.go:473 msg="total blobs: 41"
time=2026-02-17T16:09:48.102+01:00 level=INFO source=images.go:480 msg="total unused blobs removed: 0"
time=2026-02-17T16:09:48.109+01:00 level=INFO source=routes.go:1718 msg="Listening on 127.0.0.1:11434 (version 0.16.2)"
time=2026-02-17T16:09:48.124+01:00 level=INFO source=runner.go:67 msg="discovering available GPUs..."
time=2026-02-17T16:09:48.197+01:00 level=WARN source=runner.go:485 msg="user overrode visible devices" CUDA_VISIBLE_DEVICES=0
time=2026-02-17T16:09:48.197+01:00 level=WARN source=runner.go:489 msg="if GPUs are not correctly discovered, unset and try again"
time=2026-02-17T16:09:48.211+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\Users\robob\AppData\Local\Programs\Ollama\ollama.exe runner --ollama-engine --port 33827"
time=2026-02-17T16:09:48.573+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\Users\robob\AppData\Local\Programs\Ollama\ollama.exe runner --ollama-engine --port 33848"
time=2026-02-17T16:09:50.055+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\Users\robob\AppData\Local\Programs\Ollama\ollama.exe runner --ollama-engine --port 33866"
time=2026-02-17T16:09:50.457+01:00 level=INFO source=runner.go:106 msg="experimental Vulkan support disabled. To enable, set OLLAMA_VULKAN=1"
time=2026-02-17T16:09:50.460+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\Users\robob\AppData\Local\Programs\Ollama\ollama.exe runner --ollama-engine --port 33877"
time=2026-02-17T16:09:50.460+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\Users\robob\AppData\Local\Programs\Ollama\ollama.exe runner --ollama-engine --port 33878"
time=2026-02-17T16:09:50.792+01:00 level=INFO source=types.go:42 msg="inference compute" id=GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 filter_id="" library=CUDA compute=5.2 name=CUDA0 description="NVIDIA GeForce GTX 980" libdirs=ollama,cuda_v12 driver=13.0 pci_id=0000:01:00.0 type=discrete total="4.0 GiB" available="3.7 GiB"
time=2026-02-17T16:09:50.792+01:00 level=INFO source=routes.go:1768 msg="vram-based default context" total_vram="4.0 GiB" default_num_ctx=4096
[GIN] 2026/02/17 - 16:09:50 | 200 | 993.2µs | 127.0.0.1 | GET "/api/version"
[GIN] 2026/02/17 - 16:09:50 | 200 | 993.2µs | 127.0.0.1 | GET "/api/version"
[GIN] 2026/02/17 - 16:09:50 | 200 | 0s | 127.0.0.1 | GET "/api/version"
[GIN] 2026/02/17 - 16:09:50 | 200 | 7.9405ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:09:50 | 200 | 140.1508ms | 127.0.0.1 | POST "/api/show"
[GIN] 2026/02/17 - 16:09:50 | 401 | 192.2531ms | 127.0.0.1 | POST "/api/me"
[GIN] 2026/02/17 - 16:09:51 | 401 | 207.5809ms | 127.0.0.1 | POST "/api/me"
[GIN] 2026/02/17 - 16:09:57 | 200 | 112.0475ms | 127.0.0.1 | POST "/api/show"
[GIN] 2026/02/17 - 16:10:04 | 200 | 6.1967ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:10:04 | 200 | 123.9063ms | 127.0.0.1 | POST "/api/show"
[GIN] 2026/02/17 - 16:10:04 | 200 | 116.1573ms | 127.0.0.1 | POST "/api/show"
time=2026-02-17T16:10:04.333+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\Users\robob\AppData\Local\Programs\Ollama\ollama.exe runner --ollama-engine --port 33916"
time=2026-02-17T16:10:04.615+01:00 level=INFO source=cpu_windows.go:148 msg=packages count=1
time=2026-02-17T16:10:04.615+01:00 level=INFO source=cpu_windows.go:195 msg="" package=0 cores=4 efficiency=0 threads=8
llama_model_loader: loaded meta data with 26 key-value pairs and 579 tensors from F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e (version GGUF V3 (latest))
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
llama_model_loader: - kv 0: general.architecture str = qwen2
llama_model_loader: - kv 1: general.type str = model
llama_model_loader: - kv 2: general.name str = DeepSeek R1 Distill Qwen 14B
llama_model_loader: - kv 3: general.basename str = DeepSeek-R1-Distill-Qwen
llama_model_loader: - kv 4: general.size_label str = 14B
llama_model_loader: - kv 5: qwen2.block_count u32 = 48
llama_model_loader: - kv 6: qwen2.context_length u32 = 131072
llama_model_loader: - kv 7: qwen2.embedding_length u32 = 5120
llama_model_loader: - kv 8: qwen2.feed_forward_length u32 = 13824
llama_model_loader: - kv 9: qwen2.attention.head_count u32 = 40
llama_model_loader: - kv 10: qwen2.attention.head_count_kv u32 = 8
llama_model_loader: - kv 11: qwen2.rope.freq_base f32 = 1000000.000000
llama_model_loader: - kv 12: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010
llama_model_loader: - kv 13: general.file_type u32 = 15
llama_model_loader: - kv 14: tokenizer.ggml.model str = gpt2
llama_model_loader: - kv 15: tokenizer.ggml.pre str = qwen2
llama_model_loader: - kv 16: tokenizer.ggml.tokens arr[str,152064] = ["!", """, "#", "$", "%", "&", "'", ...
llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv 18: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 151646
llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151643
llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643
llama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = true
llama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false
llama_model_loader: - kv 24: tokenizer.chat_template str = {% if not add_generation_prompt is de...
llama_model_loader: - kv 25: general.quantization_version u32 = 2
llama_model_loader: - type f32: 241 tensors
llama_model_loader: - type q4_K: 289 tensors
llama_model_loader: - type q6_K: 49 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type = Q4_K - Medium
print_info: file size = 8.37 GiB (4.87 BPW)
load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
load: printing all EOG tokens:
load: - 151643 ('<｜end▁of▁sentence｜>')
load: - 151662 ('<|fim_pad|>')
load: - 151663 ('<|repo_name|>')
load: - 151664 ('<|file_sep|>')
load: special tokens cache size = 22
load: token to piece cache size = 0.9310 MB
print_info: arch = qwen2
print_info: vocab_only = 1
print_info: no_alloc = 0
print_info: model type = ?B
print_info: model params = 14.77 B
print_info: general.name = DeepSeek R1 Distill Qwen 14B
print_info: vocab type = BPE
print_info: n_vocab = 152064
print_info: n_merges = 151387
print_info: BOS token = 151646 '<｜begin▁of▁sentence｜>'
print_info: EOS token = 151643 '<｜end▁of▁sentence｜>'
print_info: EOT token = 151643 '<｜end▁of▁sentence｜>'
print_info: PAD token = 151643 '<｜end▁of▁sentence｜>'
print_info: LF token = 198 'Ċ'
print_info: FIM PRE token = 151659 '<|fim_prefix|>'
print_info: FIM SUF token = 151661 '<|fim_suffix|>'
print_info: FIM MID token = 151660 '<|fim_middle|>'
print_info: FIM PAD token = 151662 '<|fim_pad|>'
print_info: FIM REP token = 151663 '<|repo_name|>'
print_info: FIM SEP token = 151664 '<|file_sep|>'
print_info: EOG token = 151643 '<｜end▁of▁sentence｜>'
print_info: EOG token = 151662 '<|fim_pad|>'
print_info: EOG token = 151663 '<|repo_name|>'
print_info: EOG token = 151664 '<|file_sep|>'
print_info: max token length = 256
llama_model_load: vocab only - skipping tensors
time=2026-02-17T16:10:04.945+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\Users\robob\AppData\Local\Programs\Ollama\ollama.exe runner --model F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e --port 33927"
time=2026-02-17T16:10:04.946+01:00 level=INFO source=sched.go:466 msg="system memory" total="15.9 GiB" free="11.2 GiB" free_swap="19.6 GiB"
time=2026-02-17T16:10:04.946+01:00 level=INFO source=sched.go:473 msg="gpu memory" id=GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 library=CUDA available="3.3 GiB" free="3.7 GiB" minimum="457.0 MiB" overhead="0 B"
time=2026-02-17T16:10:04.946+01:00 level=INFO source=server.go:498 msg="loading model" "model layers"=49 requested=4
time=2026-02-17T16:10:04.947+01:00 level=INFO source=device.go:240 msg="model weights" device=CUDA0 size="665.7 MiB"
time=2026-02-17T16:10:04.947+01:00 level=INFO source=device.go:245 msg="model weights" device=CPU size="7.3 GiB"
time=2026-02-17T16:10:04.947+01:00 level=INFO source=device.go:251 msg="kv cache" device=CUDA0 size="64.0 MiB"
time=2026-02-17T16:10:04.947+01:00 level=INFO source=device.go:256 msg="kv cache" device=CPU size="704.0 MiB"
time=2026-02-17T16:10:04.947+01:00 level=INFO source=device.go:262 msg="compute graph" device=CUDA0 size="916.1 MiB"
time=2026-02-17T16:10:04.947+01:00 level=INFO source=device.go:272 msg="total memory" size="9.6 GiB"
time=2026-02-17T16:10:04.989+01:00 level=INFO source=runner.go:965 msg="starting go runner"
load_backend: loaded CPU backend from C:\Users\robob\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll
ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 1 CUDA devices:
Device 0: NVIDIA GeForce GTX 980, compute capability 5.2, VMM: yes, ID: GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2
load_backend: loaded CUDA backend from C:\Users\robob\AppData\Local\Programs\Ollama\lib\ollama\cuda_v12\ggml-cuda.dll
time=2026-02-17T16:10:05.177+01:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 CUDA.0.ARCHS=500,520,600,610,700,750,800,860,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(clang)
time=2026-02-17T16:10:05.178+01:00 level=INFO source=runner.go:1001 msg="Server listening on 127.0.0.1:33927"
time=2026-02-17T16:10:05.185+01:00 level=INFO source=runner.go:895 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Auto KvSize:4096 KvCacheType: NumThreads:4 GPULayers:4[ID:GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 Layers:4(44..47)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2026-02-17T16:10:05.185+01:00 level=INFO source=server.go:1350 msg="waiting for llama runner to start responding"
time=2026-02-17T16:10:05.185+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server loading model"
ggml_backend_cuda_device_get_memory device GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 utilizing NVML memory reporting free: 3986210816 total: 4294967296
llama_model_load_from_file_impl: using device CUDA0 (NVIDIA GeForce GTX 980) (0000:01:00.0) - 3801 MiB free
llama_model_loader: loaded meta data with 26 key-value pairs and 579 tensors from F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e (version GGUF V3 (latest))
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
llama_model_loader: - kv 0: general.architecture str = qwen2
llama_model_loader: - kv 1: general.type str = model
llama_model_loader: - kv 2: general.name str = DeepSeek R1 Distill Qwen 14B
llama_model_loader: - kv 3: general.basename str = DeepSeek-R1-Distill-Qwen
llama_model_loader: - kv 4: general.size_label str = 14B
llama_model_loader: - kv 5: qwen2.block_count u32 = 48
llama_model_loader: - kv 6: qwen2.context_length u32 = 131072
llama_model_loader: - kv 7: qwen2.embedding_length u32 = 5120
llama_model_loader: - kv 8: qwen2.feed_forward_length u32 = 13824
llama_model_loader: - kv 9: qwen2.attention.head_count u32 = 40
llama_model_loader: - kv 10: qwen2.attention.head_count_kv u32 = 8
llama_model_loader: - kv 11: qwen2.rope.freq_base f32 = 1000000.000000
llama_model_loader: - kv 12: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010
llama_model_loader: - kv 13: general.file_type u32 = 15
llama_model_loader: - kv 14: tokenizer.ggml.model str = gpt2
llama_model_loader: - kv 15: tokenizer.ggml.pre str = qwen2
llama_model_loader: - kv 16: tokenizer.ggml.tokens arr[str,152064] = ["!", """, "#", "$", "%", "&", "'", ...
llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv 18: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 151646
llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151643
llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643
llama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = true
llama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false
llama_model_loader: - kv 24: tokenizer.chat_template str = {% if not add_generation_prompt is de...
llama_model_loader: - kv 25: general.quantization_version u32 = 2
llama_model_loader: - type f32: 241 tensors
llama_model_loader: - type q4_K: 289 tensors
llama_model_loader: - type q6_K: 49 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type = Q4_K - Medium
print_info: file size = 8.37 GiB (4.87 BPW)
load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
load: printing all EOG tokens:
load: - 151643 ('<｜end▁of▁sentence｜>')
load: - 151662 ('<|fim_pad|>')
load: - 151663 ('<|repo_name|>')
load: - 151664 ('<|file_sep|>')
load: special tokens cache size = 22
load: token to piece cache size = 0.9310 MB
print_info: arch = qwen2
print_info: vocab_only = 0
print_info: no_alloc = 0
print_info: n_ctx_train = 131072
print_info: n_embd = 5120
print_info: n_embd_inp = 5120
print_info: n_layer = 48
print_info: n_head = 40
print_info: n_head_kv = 8
print_info: n_rot = 128
print_info: n_swa = 0
print_info: is_swa_any = 0
print_info: n_embd_head_k = 128
print_info: n_embd_head_v = 128
print_info: n_gqa = 5
print_info: n_embd_k_gqa = 1024
print_info: n_embd_v_gqa = 1024
print_info: f_norm_eps = 0.0e+00
print_info: f_norm_rms_eps = 1.0e-05
print_info: f_clamp_kqv = 0.0e+00
print_info: f_max_alibi_bias = 0.0e+00
print_info: f_logit_scale = 0.0e+00
print_info: f_attn_scale = 0.0e+00
print_info: n_ff = 13824
print_info: n_expert = 0
print_info: n_expert_used = 0
print_info: n_expert_groups = 0
print_info: n_group_used = 0
print_info: causal attn = 1
print_info: pooling type = -1
print_info: rope type = 2
print_info: rope scaling = linear
print_info: freq_base_train = 1000000.0
print_info: freq_scale_train = 1
print_info: n_ctx_orig_yarn = 131072
print_info: rope_yarn_log_mul= 0.0000
print_info: rope_finetuned = unknown
print_info: model type = 14B
print_info: model params = 14.77 B
print_info: general.name = DeepSeek R1 Distill Qwen 14B
print_info: vocab type = BPE
print_info: n_vocab = 152064
print_info: n_merges = 151387
print_info: BOS token = 151646 '<｜begin▁of▁sentence｜>'
print_info: EOS token = 151643 '<｜end▁of▁sentence｜>'
print_info: EOT token = 151643 '<｜end▁of▁sentence｜>'
print_info: PAD token = 151643 '<｜end▁of▁sentence｜>'
print_info: LF token = 198 'Ċ'
print_info: FIM PRE token = 151659 '<|fim_prefix|>'
print_info: FIM SUF token = 151661 '<|fim_suffix|>'
print_info: FIM MID token = 151660 '<|fim_middle|>'
print_info: FIM PAD token = 151662 '<|fim_pad|>'
print_info: FIM REP token = 151663 '<|repo_name|>'
print_info: FIM SEP token = 151664 '<|file_sep|>'
print_info: EOG token = 151643 '<｜end▁of▁sentence｜>'
print_info: EOG token = 151662 '<|fim_pad|>'
print_info: EOG token = 151663 '<|repo_name|>'
print_info: EOG token = 151664 '<|file_sep|>'
print_info: max token length = 256
load_tensors: loading model tensors, this can take a while... (mmap = false)
ggml_cuda_host_malloc: failed to allocate 7482.73 MiB of pinned memory: out of memory
load_tensors: offloading 4 repeating layers to GPU
load_tensors: offloaded 4/49 layers to GPU
load_tensors: CPU model buffer size = 417.66 MiB
load_tensors: CUDA0 model buffer size = 665.66 MiB
load_tensors: CPU model buffer size = 7482.73 MiB
llama_context: constructing llama_context
llama_context: n_seq_max = 1
llama_context: n_ctx = 4096
llama_context: n_ctx_seq = 4096
llama_context: n_batch = 512
llama_context: n_ubatch = 512
llama_context: causal_attn = 1
llama_context: flash_attn = auto
llama_context: kv_unified = false
llama_context: freq_base = 1000000.0
llama_context: freq_scale = 1
llama_context: n_ctx_seq (4096) < n_ctx_train (131072) -- the full capacity of the model will not be utilized
llama_context: CPU output buffer size = 0.60 MiB
llama_kv_cache: CPU KV buffer size = 704.00 MiB
llama_kv_cache: CUDA0 KV buffer size = 64.00 MiB
llama_kv_cache: size = 768.00 MiB ( 4096 cells, 48 layers, 1/1 seqs), K (f16): 384.00 MiB, V (f16): 384.00 MiB
llama_context: Flash Attention was auto, set to enabled
graph_reserve: failed to allocate compute buffers
Exception 0xc0000005 0x0 0x1f374b48f70 0x7ffceb1fe53e
PC=0x7ffceb1fe53e
signal arrived during external code execution

runtime.cgocall(0x7ff793e6e2b0, 0xc00014dc00)
runtime/cgocall.go:167 +0x3e fp=0xc00014dbd8 sp=0xc00014db70 pc=0x7ff792fd243e
github.com/ollama/ollama/llama._Cfunc_llama_init_from_model(0x1f96f41fca0, {0x1000, 0x200, 0x200, 0x1, 0x4, 0x4, 0xffffffff, 0xffffffff, 0xffffffff, ...})
_cgo_gotypes.go:759 +0x54 fp=0xc00014dc00 sp=0xc00014dbd8 pc=0x7ff793479774
github.com/ollama/ollama/llama.NewContextWithModel.func1(...)
github.com/ollama/ollama/llama/llama.go:322
github.com/ollama/ollama/llama.NewContextWithModel(0xc000346d38, {{0x1000, 0x200, 0x200, 0x1, 0x4, 0x4, 0xffffffff, 0xffffffff, 0xffffffff, ...}})
github.com/ollama/ollama/llama/llama.go:322 +0x158 fp=0xc00014dda0 sp=0xc00014dc00 pc=0x7ff79347dd58
github.com/ollama/ollama/runner/llamarunner.(*Server).loadModel(0xc0003d9b80, {{0xc0000f58f8, 0x1, 0x1}, 0x4, 0x0, 0x0, {0xc0000f58f0, 0x1, 0x2}, ...}, ...)
github.com/ollama/ollama/runner/llamarunner/runner.go:847 +0x178 fp=0xc00014dee8 sp=0xc00014dda0 pc=0x7ff79352c938
github.com/ollama/ollama/runner/llamarunner.(*Server).load.gowrap2()
github.com/ollama/ollama/runner/llamarunner/runner.go:934 +0x114 fp=0xc00014dfe0 sp=0xc00014dee8 pc=0x7ff79352db54
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc00014dfe8 sp=0xc00014dfe0 pc=0x7ff792fdd9a1
created by github.com/ollama/ollama/runner/llamarunner.(*Server).load in goroutine 10
github.com/ollama/ollama/runner/llamarunner/runner.go:934 +0x889

goroutine 1 gp=0xc0000021c0 m=nil [IO wait]:
runtime.gopark(0x7ff792fdf1a0?, 0x7ff795180d00?, 0xa0?, 0xf1?, 0xc00046f24c?)
runtime/proc.go:435 +0xce fp=0xc00011f630 sp=0xc00011f610 pc=0x7ff792fd598e
runtime.netpollblock(0x388?, 0x92f70406?, 0xf7?)
runtime/netpoll.go:575 +0xf7 fp=0xc00011f668 sp=0xc00011f630 pc=0x7ff792f9bdf7
internal/poll.runtime_pollWait(0x1f96c69a030, 0x72)
runtime/netpoll.go:351 +0x85 fp=0xc00011f688 sp=0xc00011f668 pc=0x7ff792fd4b25
internal/poll.(*pollDesc).wait(0x7ff79306a8f3?, 0x0?, 0x0)
internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc00011f6b0 sp=0xc00011f688 pc=0x7ff79306bee7
internal/poll.execIO(0xc00046f1a0, 0xc00011f758)
internal/poll/fd_windows.go:177 +0x105 fp=0xc00011f728 sp=0xc00011f6b0 pc=0x7ff79306d345
internal/poll.(*FD).acceptOne(0xc00046f188, 0x394, {0xc0001440f0?, 0xc00011f7b8?, 0x7ff793075005?}, 0xc00011f7ec?)
internal/poll/fd_windows.go:946 +0x65 fp=0xc00011f788 sp=0xc00011f728 pc=0x7ff7930718c5
internal/poll.(*FD).Accept(0xc00046f188, 0xc00011f938)
internal/poll/fd_windows.go:980 +0x1b6 fp=0xc00011f840 sp=0xc00011f788 pc=0x7ff793071bf6
net.(*netFD).accept(0xc00046f188)
net/fd_windows.go:182 +0x4b fp=0xc00011f958 sp=0xc00011f840 pc=0x7ff7930e344b
net.(*TCPListener).accept(0xc00069ad40)
net/tcpsock_posix.go:159 +0x1b fp=0xc00011f9a8 sp=0xc00011f958 pc=0x7ff7930f99fb
net.(*TCPListener).Accept(0xc00069ad40)
net/tcpsock.go:380 +0x30 fp=0xc00011f9d8 sp=0xc00011f9a8 pc=0x7ff7930f87b0
net/http.(*onceCloseListener).Accept(0xc00012c480?)
:1 +0x24 fp=0xc00011f9f0 sp=0xc00011f9d8 pc=0x7ff793311ec4
net/http.(*Server).Serve(0xc0001bfd00, {0x7ff7946e21d0, 0xc00069ad40})
net/http/server.go:3424 +0x30c fp=0xc00011fb20 sp=0xc00011f9f0 pc=0x7ff7932e978c
github.com/ollama/ollama/runner/llamarunner.Execute({0xc0000a4020, 0x4, 0x6})
github.com/ollama/ollama/runner/llamarunner/runner.go:1002 +0x8f5 fp=0xc00011fcf0 sp=0xc00011fb20 pc=0x7ff79352e515
github.com/ollama/ollama/runner.Execute({0xc0000a4010?, 0x0?, 0x0?})
github.com/ollama/ollama/runner/runner.go:25 +0x1a5 fp=0xc00011fd30 sp=0xc00011fcf0 pc=0x7ff793603ae5
github.com/ollama/ollama/cmd.NewCLI.func3(0xc0001bf800?, {0x7ff7944c4587?, 0x4?, 0x7ff7944c458b?})
github.com/ollama/ollama/cmd/cmd.go:2262 +0x45 fp=0xc00011fd58 sp=0xc00011fd30 pc=0x7ff793dfce65
github.com/spf13/cobra.(*Command).execute(0xc0001c7508, {0xc0003d3e00, 0x4, 0x4})
github.com/spf13/cobra@v1.7.0/command.go:940 +0x85c fp=0xc00011fe78 sp=0xc00011fd58 pc=0x7ff79315e61c
github.com/spf13/cobra.(*Command).ExecuteC(0xc0006d4908)
github.com/spf13/cobra@v1.7.0/command.go:1068 +0x3a5 fp=0xc00011ff30 sp=0xc00011fe78 pc=0x7ff79315ee65
github.com/spf13/cobra.(*Command).Execute(...)
github.com/spf13/cobra@v1.7.0/command.go:992
github.com/spf13/cobra.(*Command).ExecuteContext(...)
github.com/spf13/cobra@v1.7.0/command.go:985
main.main()
github.com/ollama/ollama/main.go:12 +0x4d fp=0xc00011ff50 sp=0xc00011ff30 pc=0x7ff793dff28d
runtime.main()
runtime/proc.go:283 +0x27d fp=0xc00011ffe0 sp=0xc00011ff50 pc=0x7ff792fa4ddd
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc00011ffe8 sp=0xc00011ffe0 pc=0x7ff792fdd9a1

goroutine 2 gp=0xc0000028c0 m=nil [force gc (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000071fa8 sp=0xc000071f88 pc=0x7ff792fd598e
runtime.goparkunlock(...)
runtime/proc.go:441
runtime.forcegchelper()
runtime/proc.go:348 +0xb8 fp=0xc000071fe0 sp=0xc000071fa8 pc=0x7ff792fa50f8
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000071fe8 sp=0xc000071fe0 pc=0x7ff792fdd9a1
created by runtime.init.7 in goroutine 1
runtime/proc.go:336 +0x1a

goroutine 3 gp=0xc000002c40 m=nil [GC sweep wait]:
runtime.gopark(0x1?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000073f80 sp=0xc000073f60 pc=0x7ff792fd598e
runtime.goparkunlock(...)
runtime/proc.go:441
runtime.bgsweep(0xc000080000)
runtime/mgcsweep.go:316 +0xdf fp=0xc000073fc8 sp=0xc000073f80 pc=0x7ff792f8debf
runtime.gcenable.gowrap1()
runtime/mgc.go:204 +0x25 fp=0xc000073fe0 sp=0xc000073fc8 pc=0x7ff792f82285
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000073fe8 sp=0xc000073fe0 pc=0x7ff792fdd9a1
created by runtime.gcenable in goroutine 1
runtime/mgc.go:204 +0x66

goroutine 4 gp=0xc000002e00 m=nil [GC scavenge wait]:
runtime.gopark(0x10000?, 0x7ff7946ccc38?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000087f78 sp=0xc000087f58 pc=0x7ff792fd598e
runtime.goparkunlock(...)
runtime/proc.go:441
runtime.(*scavengerState).park(0x7ff7951aa900)
runtime/mgcscavenge.go:425 +0x49 fp=0xc000087fa8 sp=0xc000087f78 pc=0x7ff792f8b909
runtime.bgscavenge(0xc000080000)
runtime/mgcscavenge.go:658 +0x59 fp=0xc000087fc8 sp=0xc000087fa8 pc=0x7ff792f8be99
runtime.gcenable.gowrap2()
runtime/mgc.go:205 +0x25 fp=0xc000087fe0 sp=0xc000087fc8 pc=0x7ff792f82225
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000087fe8 sp=0xc000087fe0 pc=0x7ff792fdd9a1
created by runtime.gcenable in goroutine 1
runtime/mgc.go:205 +0xa5

goroutine 5 gp=0xc000003340 m=nil [finalizer wait]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000089e30 sp=0xc000089e10 pc=0x7ff792fd598e
runtime.runfinq()
runtime/mfinal.go:196 +0x107 fp=0xc000089fe0 sp=0xc000089e30 pc=0x7ff792f81207
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000089fe8 sp=0xc000089fe0 pc=0x7ff792fdd9a1
created by runtime.createfing in goroutine 1
runtime/mfinal.go:166 +0x3d

goroutine 6 gp=0xc000003dc0 m=nil [chan receive]:
runtime.gopark(0xc0001f14a0?, 0xc000508018?, 0x60?, 0x5f?, 0x7ff7930cc148?)
runtime/proc.go:435 +0xce fp=0xc000075f18 sp=0xc000075ef8 pc=0x7ff792fd598e
runtime.chanrecv(0xc00008c380, 0x0, 0x1)
runtime/chan.go:664 +0x445 fp=0xc000075f90 sp=0xc000075f18 pc=0x7ff792f72d45
runtime.chanrecv1(0x7ff792fa4f40?, 0xc000075f76?)
runtime/chan.go:506 +0x12 fp=0xc000075fb8 sp=0xc000075f90 pc=0x7ff792f728d2
runtime.unique_runtime_registerUniqueMapCleanup.func2(...)
runtime/mgc.go:1796
runtime.unique_runtime_registerUniqueMapCleanup.gowrap1()
runtime/mgc.go:1799 +0x2f fp=0xc000075fe0 sp=0xc000075fb8 pc=0x7ff792f854af
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000075fe8 sp=0xc000075fe0 pc=0x7ff792fdd9a1
created by unique.runtime_registerUniqueMapCleanup in goroutine 1
runtime/mgc.go:1794 +0x85

goroutine 7 gp=0xc0003da540 m=nil [GC worker (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000083f38 sp=0xc000083f18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008d960)
runtime/mgc.go:1423 +0xe9 fp=0xc000083fc8 sp=0xc000083f38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
runtime/mgc.go:1339 +0x25 fp=0xc000083fe0 sp=0xc000083fc8 pc=0x7ff792f84685
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000083fe8 sp=0xc000083fe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
runtime/mgc.go:1339 +0x105

goroutine 18 gp=0xc0001061c0 m=nil [GC worker (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000113f38 sp=0xc000113f18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008d960)
runtime/mgc.go:1423 +0xe9 fp=0xc000113fc8 sp=0xc000113f38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
runtime/mgc.go:1339 +0x25 fp=0xc000113fe0 sp=0xc000113fc8 pc=0x7ff792f84685
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000113fe8 sp=0xc000113fe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
runtime/mgc.go:1339 +0x105

goroutine 34 gp=0xc000484000 m=nil [GC worker (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc00010ff38 sp=0xc00010ff18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008d960)
runtime/mgc.go:1423 +0xe9 fp=0xc00010ffc8 sp=0xc00010ff38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
runtime/mgc.go:1339 +0x25 fp=0xc00010ffe0 sp=0xc00010ffc8 pc=0x7ff792f84685
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc00010ffe8 sp=0xc00010ffe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
runtime/mgc.go:1339 +0x105

goroutine 35 gp=0xc0004841c0 m=nil [GC worker (idle)]:
runtime.gopark(0x4b423cff1788?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000111f38 sp=0xc000111f18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008d960)
runtime/mgc.go:1423 +0xe9 fp=0xc000111fc8 sp=0xc000111f38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
runtime/mgc.go:1339 +0x25 fp=0xc000111fe0 sp=0xc000111fc8 pc=0x7ff792f84685
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000111fe8 sp=0xc000111fe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
runtime/mgc.go:1339 +0x105

goroutine 36 gp=0xc000484380 m=nil [GC worker (idle)]:
runtime.gopark(0x4b423cf7000c?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc00048bf38 sp=0xc00048bf18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008d960)
runtime/mgc.go:1423 +0xe9 fp=0xc00048bfc8 sp=0xc00048bf38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
runtime/mgc.go:1339 +0x25 fp=0xc00048bfe0 sp=0xc00048bfc8 pc=0x7ff792f84685
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc00048bfe8 sp=0xc00048bfe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
runtime/mgc.go:1339 +0x105

goroutine 8 gp=0xc0003da700 m=nil [GC worker (idle)]:
runtime.gopark(0x4b423cff1788?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000085f38 sp=0xc000085f18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008d960)
runtime/mgc.go:1423 +0xe9 fp=0xc000085fc8 sp=0xc000085f38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
runtime/mgc.go:1339 +0x25 fp=0xc000085fe0 sp=0xc000085fc8 pc=0x7ff792f84685
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000085fe8 sp=0xc000085fe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
runtime/mgc.go:1339 +0x105

goroutine 19 gp=0xc000106380 m=nil [GC worker (idle)]:
runtime.gopark(0x4b423cff1788?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000115f38 sp=0xc000115f18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008d960)
runtime/mgc.go:1423 +0xe9 fp=0xc000115fc8 sp=0xc000115f38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
runtime/mgc.go:1339 +0x25 fp=0xc000115fe0 sp=0xc000115fc8 pc=0x7ff792f84685
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000115fe8 sp=0xc000115fe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
runtime/mgc.go:1339 +0x105

goroutine 37 gp=0xc000484540 m=nil [GC worker (idle)]:
runtime.gopark(0x7ff7951fc600?, 0x1?, 0x48?, 0x5c?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc00048df38 sp=0xc00048df18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008d960)
runtime/mgc.go:1423 +0xe9 fp=0xc00048dfc8 sp=0xc00048df38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
runtime/mgc.go:1339 +0x25 fp=0xc00048dfe0 sp=0xc00048dfc8 pc=0x7ff792f84685
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc00048dfe8 sp=0xc00048dfe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
runtime/mgc.go:1339 +0x105

goroutine 9 gp=0xc000484c40 m=nil [sync.WaitGroup.Wait]:
runtime.gopark(0x0?, 0x0?, 0x60?, 0x3e?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000487e20 sp=0xc000487e00 pc=0x7ff792fd598e
runtime.goparkunlock(...)
runtime/proc.go:441
runtime.semacquire1(0xc0003d9ba0, 0x0, 0x1, 0x0, 0x18)
runtime/sema.go:188 +0x22f fp=0xc000487e88 sp=0xc000487e20 pc=0x7ff792fb750f
sync.runtime_SemacquireWaitGroup(0x0?)
runtime/sema.go:110 +0x25 fp=0xc000487ec0 sp=0xc000487e88 pc=0x7ff792fd6f85
sync.(*WaitGroup).Wait(0x0?)
sync/waitgroup.go:118 +0x48 fp=0xc000487ee8 sp=0xc000487ec0 pc=0x7ff792feb988
github.com/ollama/ollama/runner/llamarunner.(*Server).run(0xc0003d9b80, {0x7ff7946e4a30, 0xc00033f770})
github.com/ollama/ollama/runner/llamarunner/runner.go:360 +0x4b fp=0xc000487fb8 sp=0xc000487ee8 pc=0x7ff79352922b
github.com/ollama/ollama/runner/llamarunner.Execute.gowrap1()
github.com/ollama/ollama/runner/llamarunner/runner.go:981 +0x28 fp=0xc000487fe0 sp=0xc000487fb8 pc=0x7ff79352e788
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000487fe8 sp=0xc000487fe0 pc=0x7ff792fdd9a1
created by github.com/ollama/ollama/runner/llamarunner.Execute in goroutine 1
github.com/ollama/ollama/runner/llamarunner/runner.go:981 +0x4c5

goroutine 10 gp=0xc000484e00 m=nil [IO wait]:
runtime.gopark(0x0?, 0xc00046f420?, 0xc8?, 0xf4?, 0xc00046f4cc?)
runtime/proc.go:435 +0xce fp=0xc00004b8c8 sp=0xc00004b8a8 pc=0x7ff792fd598e
runtime.netpollblock(0x390?, 0x92f70406?, 0xf7?)
runtime/netpoll.go:575 +0xf7 fp=0xc00004b900 sp=0xc00004b8c8 pc=0x7ff792f9bdf7
internal/poll.runtime_pollWait(0x1f96c699f18, 0x72)
runtime/netpoll.go:351 +0x85 fp=0xc00004b920 sp=0xc00004b900 pc=0x7ff792fd4b25
internal/poll.(*pollDesc).wait(0x7ff79319d5f7?, 0xc00004b970?, 0x0)
internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc00004b948 sp=0xc00004b920 pc=0x7ff79306bee7
internal/poll.execIO(0xc00046f420, 0x7ff794546098)
internal/poll/fd_windows.go:177 +0x105 fp=0xc00004b9c0 sp=0xc00004b948 pc=0x7ff79306d345
internal/poll.(*FD).Read(0xc00046f408, {0xc000091000, 0x1000, 0x1000})
internal/poll/fd_windows.go:438 +0x29b fp=0xc00004ba60 sp=0xc00004b9c0 pc=0x7ff79306e01b
net.(*netFD).Read(0xc00046f408, {0xc000091000?, 0xc00004bad0?, 0x7ff79306c3a5?})
net/fd_posix.go:55 +0x25 fp=0xc00004baa8 sp=0xc00004ba60 pc=0x7ff7930e1325
net.(*conn).Read(0xc000078740, {0xc000091000?, 0x0?, 0x0?})
net/net.go:194 +0x45 fp=0xc00004baf0 sp=0xc00004baa8 pc=0x7ff7930f0a45
net/http.(*connReader).Read(0xc000246090, {0xc000091000, 0x1000, 0x1000})
net/http/server.go:798 +0x159 fp=0xc00004bb40 sp=0xc00004baf0 pc=0x7ff7932de639
bufio.(*Reader).fill(0xc000213ec0)
bufio/bufio.go:113 +0x103 fp=0xc00004bb78 sp=0xc00004bb40 pc=0x7ff793107963
bufio.(*Reader).Peek(0xc000213ec0, 0x4)
bufio/bufio.go:152 +0x53 fp=0xc00004bb98 sp=0xc00004bb78 pc=0x7ff793107a93
net/http.(*conn).serve(0xc00012c480, {0x7ff7946e49f8, 0xc0003243f0})
net/http/server.go:2137 +0x785 fp=0xc00004bfb8 sp=0xc00004bb98 pc=0x7ff7932e4425
net/http.(*Server).Serve.gowrap3()
net/http/server.go:3454 +0x28 fp=0xc00004bfe0 sp=0xc00004bfb8 pc=0x7ff7932e9b88
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc00004bfe8 sp=0xc00004bfe0 pc=0x7ff792fdd9a1
created by net/http.(*Server).Serve in goroutine 1
net/http/server.go:3454 +0x485
rax 0xfffffff9fdb6ed00
rbx 0x1f976fda270
rcx 0x1f96f61a8a0
rdx 0x0
rdi 0x0
rsi 0x1f96f61cc80
rbp 0xf1506ffbd0
rsp 0xf1506ff920
r8 0x1f947061080
r9 0x1
r10 0x8000
r11 0xf1506ff930
r12 0x667
r13 0x1218
r14 0x1
r15 0x1f96f41fca0
rip 0x7ffceb1fe53e
rflags 0x10287
cs 0x33
fs 0x53
gs 0x2b
time=2026-02-17T16:10:27.322+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server error"
time=2026-02-17T16:10:28.074+01:00 level=INFO source=sched.go:493 msg="Load failed" model=F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e error="llama runner process has terminated: exit status 2"
[GIN] 2026/02/17 - 16:10:28 | 500 | 23.8713951s | 127.0.0.1 | POST "/api/chat"
[GIN] 2026/02/17 - 16:10:34 | 200 | 12.5133ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:11:04 | 200 | 5.2293ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:11:34 | 200 | 4.2837ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:12:04 | 200 | 12.7989ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:12:34 | 200 | 4.2427ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:13:04 | 200 | 5.9933ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:13:34 | 200 | 5.0279ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:14:04 | 200 | 11.724ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:14:34 | 200 | 4.7326ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:15:04 | 200 | 5.4637ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:15:34 | 200 | 4.426ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:16:04 | 200 | 5.009ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:16:34 | 200 | 4.6721ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:17:04 | 200 | 5.6033ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:17:34 | 200 | 4.5266ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:18:04 | 200 | 4.8485ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:18:34 | 200 | 4.644ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:18:59 | 200 | 0s | 127.0.0.1 | HEAD "/"
[GIN] 2026/02/17 - 16:19:05 | 200 | 5.0163ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:19:36 | 200 | 4.1555ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:19:40 | 200 | 0s | 127.0.0.1 | HEAD "/"
[GIN] 2026/02/17 - 16:19:40 | 200 | 52.8589ms | 127.0.0.1 | POST "/api/create"
[GIN] 2026/02/17 - 16:19:51 | 200 | 0s | 127.0.0.1 | HEAD "/"
[GIN] 2026/02/17 - 16:19:51 | 200 | 46.3741ms | 127.0.0.1 | POST "/api/create"
[GIN] 2026/02/17 - 16:20:07 | 200 | 5.1281ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:20:08 | 200 | 0s | 127.0.0.1 | HEAD "/"
[GIN] 2026/02/17 - 16:20:08 | 200 | 38.7198ms | 127.0.0.1 | POST "/api/create"
[GIN] 2026/02/17 - 16:20:11 | 200 | 1.0024ms | 127.0.0.1 | GET "/api/version"
[GIN] 2026/02/17 - 16:20:11 | 200 | 16.2654ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:20:11 | 401 | 192.0709ms | 127.0.0.1 | POST "/api/me"
[GIN] 2026/02/17 - 16:20:13 | 200 | 123.5651ms | 127.0.0.1 | POST "/api/show"
[GIN] 2026/02/17 - 16:20:16 | 200 | 120.6338ms | 127.0.0.1 | POST "/api/show"
[GIN] 2026/02/17 - 16:20:16 | 200 | 116.2417ms | 127.0.0.1 | POST "/api/show"
time=2026-02-17T16:20:17.006+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\Users\robob\AppData\Local\Programs\Ollama\ollama.exe runner --ollama-engine --port 34051"
time=2026-02-17T16:20:17.283+01:00 level=INFO source=cpu_windows.go:148 msg=packages count=1
time=2026-02-17T16:20:17.283+01:00 level=INFO source=cpu_windows.go:195 msg="" package=0 cores=4 efficiency=0 threads=8
llama_model_loader: loaded meta data with 26 key-value pairs and 579 tensors from F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e (version GGUF V3 (latest))
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
llama_model_loader: - kv 0: general.architecture str = qwen2
llama_model_loader: - kv 1: general.type str = model
llama_model_loader: - kv 2: general.name str = DeepSeek R1 Distill Qwen 14B
llama_model_loader: - kv 3: general.basename str = DeepSeek-R1-Distill-Qwen
llama_model_loader: - kv 4: general.size_label str = 14B
llama_model_loader: - kv 5: qwen2.block_count u32 = 48
llama_model_loader: - kv 6: qwen2.context_length u32 = 131072
llama_model_loader: - kv 7: qwen2.embedding_length u32 = 5120
llama_model_loader: - kv 8: qwen2.feed_forward_length u32 = 13824
llama_model_loader: - kv 9: qwen2.attention.head_count u32 = 40
llama_model_loader: - kv 10: qwen2.attention.head_count_kv u32 = 8
llama_model_loader: - kv 11: qwen2.rope.freq_base f32 = 1000000.000000
llama_model_loader: - kv 12: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010
llama_model_loader: - kv 13: general.file_type u32 = 15
llama_model_loader: - kv 14: tokenizer.ggml.model str = gpt2
llama_model_loader: - kv 15: tokenizer.ggml.pre str = qwen2
llama_model_loader: - kv 16: tokenizer.ggml.tokens arr[str,152064] = ["!", """, "#", "$", "%", "&", "'", ...
llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv 18: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 151646
llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151643
llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643
llama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = true
llama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false
llama_model_loader: - kv 24: tokenizer.chat_template str = {% if not add_generation_prompt is de...
llama_model_loader: - kv 25: general.quantization_version u32 = 2
llama_model_loader: - type f32: 241 tensors
llama_model_loader: - type q4_K: 289 tensors
llama_model_loader: - type q6_K: 49 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type = Q4_K - Medium
print_info: file size = 8.37 GiB (4.87 BPW)
load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
load: printing all EOG tokens:
load: - 151643 ('<｜end▁of▁sentence｜>')
load: - 151662 ('<|fim_pad|>')
load: - 151663 ('<|repo_name|>')
load: - 151664 ('<|file_sep|>')
load: special tokens cache size = 22
load: token to piece cache size = 0.9310 MB
print_info: arch = qwen2
print_info: vocab_only = 1
print_info: no_alloc = 0
print_info: model type = ?B
print_info: model params = 14.77 B
print_info: general.name = DeepSeek R1 Distill Qwen 14B
print_info: vocab type = BPE
print_info: n_vocab = 152064
print_info: n_merges = 151387
print_info: BOS token = 151646 '<｜begin▁of▁sentence｜>'
print_info: EOS token = 151643 '<｜end▁of▁sentence｜>'
print_info: EOT token = 151643 '<｜end▁of▁sentence｜>'
print_info: PAD token = 151643 '<｜end▁of▁sentence｜>'
print_info: LF token = 198 'Ċ'
print_info: FIM PRE token = 151659 '<|fim_prefix|>'
print_info: FIM SUF token = 151661 '<|fim_suffix|>'
print_info: FIM MID token = 151660 '<|fim_middle|>'
print_info: FIM PAD token = 151662 '<|fim_pad|>'
print_info: FIM REP token = 151663 '<|repo_name|>'
print_info: FIM SEP token = 151664 '<|file_sep|>'
print_info: EOG token = 151643 '<｜end▁of▁sentence｜>'
print_info: EOG token = 151662 '<|fim_pad|>'
print_info: EOG token = 151663 '<|repo_name|>'
print_info: EOG token = 151664 '<|file_sep|>'
print_info: max token length = 256
llama_model_load: vocab only - skipping tensors
time=2026-02-17T16:20:17.602+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\Users\robob\AppData\Local\Programs\Ollama\ollama.exe runner --model F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e --port 34062"
time=2026-02-17T16:20:17.603+01:00 level=INFO source=sched.go:466 msg="system memory" total="15.9 GiB" free="9.1 GiB" free_swap="16.7 GiB"
time=2026-02-17T16:20:17.603+01:00 level=INFO source=sched.go:473 msg="gpu memory" id=GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 library=CUDA available="2.9 GiB" free="3.3 GiB" minimum="457.0 MiB" overhead="0 B"
time=2026-02-17T16:20:17.603+01:00 level=INFO source=server.go:498 msg="loading model" "model layers"=49 requested=2
time=2026-02-17T16:20:17.604+01:00 level=INFO source=device.go:240 msg="model weights" device=CUDA0 size="332.8 MiB"
time=2026-02-17T16:20:17.604+01:00 level=INFO source=device.go:245 msg="model weights" device=CPU size="7.6 GiB"
time=2026-02-17T16:20:17.604+01:00 level=INFO source=device.go:251 msg="kv cache" device=CUDA0 size="32.0 MiB"
time=2026-02-17T16:20:17.604+01:00 level=INFO source=device.go:256 msg="kv cache" device=CPU size="736.0 MiB"
time=2026-02-17T16:20:17.604+01:00 level=INFO source=device.go:262 msg="compute graph" device=CUDA0 size="916.1 MiB"
time=2026-02-17T16:20:17.604+01:00 level=INFO source=device.go:272 msg="total memory" size="9.6 GiB"
time=2026-02-17T16:20:17.646+01:00 level=INFO source=runner.go:965 msg="starting go runner"
load_backend: loaded CPU backend from C:\Users\robob\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll
ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 1 CUDA devices:
Device 0: NVIDIA GeForce GTX 980, compute capability 5.2, VMM: yes, ID: GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2
load_backend: loaded CUDA backend from C:\Users\robob\AppData\Local\Programs\Ollama\lib\ollama\cuda_v12\ggml-cuda.dll
time=2026-02-17T16:20:17.835+01:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 CUDA.0.ARCHS=500,520,600,610,700,750,800,860,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(clang)
time=2026-02-17T16:20:17.836+01:00 level=INFO source=runner.go:1001 msg="Server listening on 127.0.0.1:34062"
time=2026-02-17T16:20:17.844+01:00 level=INFO source=runner.go:895 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Auto KvSize:4096 KvCacheType: NumThreads:4 GPULayers:2[ID:GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 Layers:2(46..47)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2026-02-17T16:20:17.844+01:00 level=INFO source=server.go:1350 msg="waiting for llama runner to start responding"
time=2026-02-17T16:20:17.844+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server loading model"
ggml_backend_cuda_device_get_memory device GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 utilizing NVML memory reporting free: 3549564928 total: 4294967296
llama_model_load_from_file_impl: using device CUDA0 (NVIDIA GeForce GTX 980) (0000:01:00.0) - 3385 MiB free
llama_model_loader: loaded meta data with 26 key-value pairs and 579 tensors from F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e (version GGUF V3 (latest))
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
llama_model_loader: - kv 0: general.architecture str = qwen2
llama_model_loader: - kv 1: general.type str = model
llama_model_loader: - kv 2: general.name str = DeepSeek R1 Distill Qwen 14B
llama_model_loader: - kv 3: general.basename str = DeepSeek-R1-Distill-Qwen
llama_model_loader: - kv 4: general.size_label str = 14B
llama_model_loader: - kv 5: qwen2.block_count u32 = 48
llama_model_loader: - kv 6: qwen2.context_length u32 = 131072
llama_model_loader: - kv 7: qwen2.embedding_length u32 = 5120
llama_model_loader: - kv 8: qwen2.feed_forward_length u32 = 13824
llama_model_loader: - kv 9: qwen2.attention.head_count u32 = 40
llama_model_loader: - kv 10: qwen2.attention.head_count_kv u32 = 8
llama_model_loader: - kv 11: qwen2.rope.freq_base f32 = 1000000.000000
llama_model_loader: - kv 12: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010
llama_model_loader: - kv 13: general.file_type u32 = 15
llama_model_loader: - kv 14: tokenizer.ggml.model str = gpt2
llama_model_loader: - kv 15: tokenizer.ggml.pre str = qwen2
llama_model_loader: - kv 16: tokenizer.ggml.tokens arr[str,152064] = ["!", """, "#", "$", "%", "&", "'", ...
llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv 18: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 151646
llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151643
llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643
llama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = true
llama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false
llama_model_loader: - kv 24: tokenizer.chat_template str = {% if not add_generation_prompt is de...
llama_model_loader: - kv 25: general.quantization_version u32 = 2
llama_model_loader: - type f32: 241 tensors
llama_model_loader: - type q4_K: 289 tensors
llama_model_loader: - type q6_K: 49 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type = Q4_K - Medium
print_info: file size = 8.37 GiB (4.87 BPW)
load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
load: printing all EOG tokens:
load: - 151643 ('<｜end▁of▁sentence｜>')
load: - 151662 ('<|fim_pad|>')
load: - 151663 ('<|repo_name|>')
load: - 151664 ('<|file_sep|>')
load: special tokens cache size = 22
load: token to piece cache size = 0.9310 MB
print_info: arch = qwen2
print_info: vocab_only = 0
print_info: no_alloc = 0
print_info: n_ctx_train = 131072
print_info: n_embd = 5120
print_info: n_embd_inp = 5120
print_info: n_layer = 48
print_info: n_head = 40
print_info: n_head_kv = 8
print_info: n_rot = 128
print_info: n_swa = 0
print_info: is_swa_any = 0
print_info: n_embd_head_k = 128
print_info: n_embd_head_v = 128
print_info: n_gqa = 5
print_info: n_embd_k_gqa = 1024
print_info: n_embd_v_gqa = 1024
print_info: f_norm_eps = 0.0e+00
print_info: f_norm_rms_eps = 1.0e-05
print_info: f_clamp_kqv = 0.0e+00
print_info: f_max_alibi_bias = 0.0e+00
print_info: f_logit_scale = 0.0e+00
print_info: f_attn_scale = 0.0e+00
print_info: n_ff = 13824
print_info: n_expert = 0
print_info: n_expert_used = 0
print_info: n_expert_groups = 0
print_info: n_group_used = 0
print_info: causal attn = 1
print_info: pooling type = -1
print_info: rope type = 2
print_info: rope scaling = linear
print_info: freq_base_train = 1000000.0
print_info: freq_scale_train = 1
print_info: n_ctx_orig_yarn = 131072
print_info: rope_yarn_log_mul= 0.0000
print_info: rope_finetuned = unknown
print_info: model type = 14B
print_info: model params = 14.77 B
print_info: general.name = DeepSeek R1 Distill Qwen 14B
print_info: vocab type = BPE
print_info: n_vocab = 152064
print_info: n_merges = 151387
print_info: BOS token = 151646 '<｜begin▁of▁sentence｜>'
print_info: EOS token = 151643 '<｜end▁of▁sentence｜>'
print_info: EOT token = 151643 '<｜end▁of▁sentence｜>'
print_info: PAD token = 151643 '<｜end▁of▁sentence｜>'
print_info: LF token = 198 'Ċ'
print_info: FIM PRE token = 151659 '<|fim_prefix|>'
print_info: FIM SUF token = 151661 '<|fim_suffix|>'
print_info: FIM MID token = 151660 '<|fim_middle|>'
print_info: FIM PAD token = 151662 '<|fim_pad|>'
print_info: FIM REP token = 151663 '<|repo_name|>'
print_info: FIM SEP token = 151664 '<|file_sep|>'
print_info: EOG token = 151643 '<｜end▁of▁sentence｜>'
print_info: EOG token = 151662 '<|fim_pad|>'
print_info: EOG token = 151663 '<|repo_name|>'
print_info: EOG token = 151664 '<|file_sep|>'
print_info: max token length = 256
load_tensors: loading model tensors, this can take a while... (mmap = false)
ggml_cuda_host_malloc: failed to allocate 7815.55 MiB of pinned memory: out of memory
load_tensors: offloading 2 repeating layers to GPU
load_tensors: offloaded 2/49 layers to GPU
load_tensors: CPU model buffer size = 417.66 MiB
load_tensors: CUDA0 model buffer size = 332.83 MiB
load_tensors: CPU model buffer size = 7815.55 MiB
llama_context: constructing llama_context
llama_context: n_seq_max = 1
llama_context: n_ctx = 4096
llama_context: n_ctx_seq = 4096
llama_context: n_batch = 512
llama_context: n_ubatch = 512
llama_context: causal_attn = 1
llama_context: flash_attn = auto
llama_context: kv_unified = false
llama_context: freq_base = 1000000.0
llama_context: freq_scale = 1
llama_context: n_ctx_seq (4096) < n_ctx_train (131072) -- the full capacity of the model will not be utilized
llama_context: CPU output buffer size = 0.60 MiB
llama_kv_cache: CPU KV buffer size = 736.00 MiB
[GIN] 2026/02/17 - 16:20:41 | 200 | 14.9989ms | 127.0.0.1 | GET "/api/tags"
llama_kv_cache: CUDA0 KV buffer size = 32.00 MiB
llama_kv_cache: size = 768.00 MiB ( 4096 cells, 48 layers, 1/1 seqs), K (f16): 384.00 MiB, V (f16): 384.00 MiB
llama_context: Flash Attention was auto, set to enabled
graph_reserve: failed to allocate compute buffers
Exception 0xc0000005 0x0 0x243fe0a9d60 0x7ffceb1fe53e
PC=0x7ffceb1fe53e
signal arrived during external code execution

runtime.cgocall(0x7ff793e6e2b0, 0xc000365c00)
runtime/cgocall.go:167 +0x3e fp=0xc000365bd8 sp=0xc000365b70 pc=0x7ff792fd243e
github.com/ollama/ollama/llama._Cfunc_llama_init_from_model(0x258f4221790, {0x1000, 0x200, 0x200, 0x1, 0x4, 0x4, 0xffffffff, 0xffffffff, 0xffffffff, ...})
_cgo_gotypes.go:759 +0x54 fp=0xc000365c00 sp=0xc000365bd8 pc=0x7ff793479774
github.com/ollama/ollama/llama.NewContextWithModel.func1(...)
github.com/ollama/ollama/llama/llama.go:322
github.com/ollama/ollama/llama.NewContextWithModel(0xc0001cc020, {{0x1000, 0x200, 0x200, 0x1, 0x4, 0x4, 0xffffffff, 0xffffffff, 0xffffffff, ...}})
github.com/ollama/ollama/llama/llama.go:322 +0x158 fp=0xc000365da0 sp=0xc000365c00 pc=0x7ff79347dd58
github.com/ollama/ollama/runner/llamarunner.(*Server).loadModel(0xc0003db0e0, {{0xc0000f58f8, 0x1, 0x1}, 0x2, 0x0, 0x0, {0xc0000f58f0, 0x1, 0x2}, ...}, ...)
github.com/ollama/ollama/runner/llamarunner/runner.go:847 +0x178 fp=0xc000365ee8 sp=0xc000365da0 pc=0x7ff79352c938
github.com/ollama/ollama/runner/llamarunner.(*Server).load.gowrap2()
github.com/ollama/ollama/runner/llamarunner/runner.go:934 +0x114 fp=0xc000365fe0 sp=0xc000365ee8 pc=0x7ff79352db54
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000365fe8 sp=0xc000365fe0 pc=0x7ff792fdd9a1
created by github.com/ollama/ollama/runner/llamarunner.(*Server).load in goroutine 12
github.com/ollama/ollama/runner/llamarunner/runner.go:934 +0x889

goroutine 1 gp=0xc0000021c0 m=nil [IO wait]:
runtime.gopark(0x7ff792fdf1a0?, 0x7ff795180d00?, 0xa0?, 0xf1?, 0xc00047f24c?)
runtime/proc.go:435 +0xce fp=0xc000587630 sp=0xc000587610 pc=0x7ff792fd598e
runtime.netpollblock(0x380?, 0x92f70406?, 0xf7?)
runtime/netpoll.go:575 +0xf7 fp=0xc000587668 sp=0xc000587630 pc=0x7ff792f9bdf7
internal/poll.runtime_pollWait(0x258f1616050, 0x72)
runtime/netpoll.go:351 +0x85 fp=0xc000587688 sp=0xc000587668 pc=0x7ff792fd4b25
internal/poll.(*pollDesc).wait(0x7ff79306a8f3?, 0x0?, 0x0)
internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc0005876b0 sp=0xc000587688 pc=0x7ff79306bee7
internal/poll.execIO(0xc00047f1a0, 0xc000587758)
internal/poll/fd_windows.go:177 +0x105 fp=0xc000587728 sp=0xc0005876b0 pc=0x7ff79306d345
internal/poll.(*FD).acceptOne(0xc00047f188, 0x38c, {0xc0003560f0?, 0xc0005877b8?, 0x7ff793075005?}, 0xc0005877ec?)
internal/poll/fd_windows.go:946 +0x65 fp=0xc000587788 sp=0xc000587728 pc=0x7ff7930718c5
internal/poll.(*FD).Accept(0xc00047f188, 0xc000587938)
internal/poll/fd_windows.go:980 +0x1b6 fp=0xc000587840 sp=0xc000587788 pc=0x7ff793071bf6
net.(*netFD).accept(0xc00047f188)
net/fd_windows.go:182 +0x4b fp=0xc000587958 sp=0xc000587840 pc=0x7ff7930e344b
net.(*TCPListener).accept(0xc0002ae500)
net/tcpsock_posix.go:159 +0x1b fp=0xc0005879a8 sp=0xc000587958 pc=0x7ff7930f99fb
net.(*TCPListener).Accept(0xc0002ae500)
net/tcpsock.go:380 +0x30 fp=0xc0005879d8 sp=0xc0005879a8 pc=0x7ff7930f87b0
net/http.(*onceCloseListener).Accept(0xc000204480?)
:1 +0x24 fp=0xc0005879f0 sp=0xc0005879d8 pc=0x7ff793311ec4
net/http.(*Server).Serve(0xc00013fd00, {0x7ff7946e21d0, 0xc0002ae500})
net/http/server.go:3424 +0x30c fp=0xc000587b20 sp=0xc0005879f0 pc=0x7ff7932e978c
github.com/ollama/ollama/runner/llamarunner.Execute({0xc0000a4020, 0x4, 0x6})
github.com/ollama/ollama/runner/llamarunner/runner.go:1002 +0x8f5 fp=0xc000587cf0 sp=0xc000587b20 pc=0x7ff79352e515
github.com/ollama/ollama/runner.Execute({0xc0000a4010?, 0x0?, 0x0?})
github.com/ollama/ollama/runner/runner.go:25 +0x1a5 fp=0xc000587d30 sp=0xc000587cf0 pc=0x7ff793603ae5
github.com/ollama/ollama/cmd.NewCLI.func3(0xc00013f800?, {0x7ff7944c4587?, 0x4?, 0x7ff7944c458b?})
github.com/ollama/ollama/cmd/cmd.go:2262 +0x45 fp=0xc000587d58 sp=0xc000587d30 pc=0x7ff793dfce65
github.com/spf13/cobra.(*Command).execute(0xc0002d3b08, {0xc0002ae280, 0x4, 0x4})
github.com/spf13/cobra@v1.7.0/command.go:940 +0x85c fp=0xc000587e78 sp=0xc000587d58 pc=0x7ff79315e61c
github.com/spf13/cobra.(*Command).ExecuteC(0xc0004d6f08)
github.com/spf13/cobra@v1.7.0/command.go:1068 +0x3a5 fp=0xc000587f30 sp=0xc000587e78 pc=0x7ff79315ee65
github.com/spf13/cobra.(*Command).Execute(...)
github.com/spf13/cobra@v1.7.0/command.go:992
github.com/spf13/cobra.(*Command).ExecuteContext(...)
github.com/spf13/cobra@v1.7.0/command.go:985
main.main()
github.com/ollama/ollama/main.go:12 +0x4d fp=0xc000587f50 sp=0xc000587f30 pc=0x7ff793dff28d
runtime.main()
runtime/proc.go:283 +0x27d fp=0xc000587fe0 sp=0xc000587f50 pc=0x7ff792fa4ddd
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000587fe8 sp=0xc000587fe0 pc=0x7ff792fdd9a1

goroutine 2 gp=0xc0000028c0 m=nil [force gc (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000071fa8 sp=0xc000071f88 pc=0x7ff792fd598e
runtime.goparkunlock(...)
runtime/proc.go:441
runtime.forcegchelper()
runtime/proc.go:348 +0xb8 fp=0xc000071fe0 sp=0xc000071fa8 pc=0x7ff792fa50f8
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000071fe8 sp=0xc000071fe0 pc=0x7ff792fdd9a1
created by runtime.init.7 in goroutine 1
runtime/proc.go:336 +0x1a

goroutine 3 gp=0xc000002c40 m=nil [GC sweep wait]:
runtime.gopark(0x1?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000073f80 sp=0xc000073f60 pc=0x7ff792fd598e
runtime.goparkunlock(...)
runtime/proc.go:441
runtime.bgsweep(0xc000080000)
runtime/mgcsweep.go:316 +0xdf fp=0xc000073fc8 sp=0xc000073f80 pc=0x7ff792f8debf
runtime.gcenable.gowrap1()
runtime/mgc.go:204 +0x25 fp=0xc000073fe0 sp=0xc000073fc8 pc=0x7ff792f82285
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000073fe8 sp=0xc000073fe0 pc=0x7ff792fdd9a1
created by runtime.gcenable in goroutine 1
runtime/mgc.go:204 +0x66

goroutine 4 gp=0xc000002e00 m=nil [GC scavenge wait]:
runtime.gopark(0x10000?, 0x7ff7946ccc38?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000087f78 sp=0xc000087f58 pc=0x7ff792fd598e
runtime.goparkunlock(...)
runtime/proc.go:441
runtime.(*scavengerState).park(0x7ff7951aa900)
runtime/mgcscavenge.go:425 +0x49 fp=0xc000087fa8 sp=0xc000087f78 pc=0x7ff792f8b909
runtime.bgscavenge(0xc000080000)
runtime/mgcscavenge.go:658 +0x59 fp=0xc000087fc8 sp=0xc000087fa8 pc=0x7ff792f8be99
runtime.gcenable.gowrap2()
runtime/mgc.go:205 +0x25 fp=0xc000087fe0 sp=0xc000087fc8 pc=0x7ff792f82225
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000087fe8 sp=0xc000087fe0 pc=0x7ff792fdd9a1
created by runtime.gcenable in goroutine 1
runtime/mgc.go:205 +0xa5

goroutine 5 gp=0xc000003340 m=nil [finalizer wait]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000089e30 sp=0xc000089e10 pc=0x7ff792fd598e
runtime.runfinq()
runtime/mfinal.go:196 +0x107 fp=0xc000089fe0 sp=0xc000089e30 pc=0x7ff792f81207
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000089fe8 sp=0xc000089fe0 pc=0x7ff792fdd9a1
created by runtime.createfing in goroutine 1
runtime/mfinal.go:166 +0x3d

goroutine 6 gp=0xc000003dc0 m=nil [chan receive]:
runtime.gopark(0xc00017b2c0?, 0xc000508018?, 0x60?, 0x5f?, 0x7ff7930cc148?)
runtime/proc.go:435 +0xce fp=0xc000075f18 sp=0xc000075ef8 pc=0x7ff792fd598e
runtime.chanrecv(0xc00008c380, 0x0, 0x1)
runtime/chan.go:664 +0x445 fp=0xc000075f90 sp=0xc000075f18 pc=0x7ff792f72d45
runtime.chanrecv1(0x7ff792fa4f40?, 0xc000075f76?)
runtime/chan.go:506 +0x12 fp=0xc000075fb8 sp=0xc000075f90 pc=0x7ff792f728d2
runtime.unique_runtime_registerUniqueMapCleanup.func2(...)
runtime/mgc.go:1796
runtime.unique_runtime_registerUniqueMapCleanup.gowrap1()
runtime/mgc.go:1799 +0x2f fp=0xc000075fe0 sp=0xc000075fb8 pc=0x7ff792f854af
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000075fe8 sp=0xc000075fe0 pc=0x7ff792fdd9a1
created by unique.runtime_registerUniqueMapCleanup in goroutine 1
runtime/mgc.go:1794 +0x85

goroutine 7 gp=0xc0003d8540 m=nil [GC worker (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000083f38 sp=0xc000083f18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008d960)
runtime/mgc.go:1423 +0xe9 fp=0xc000083fc8 sp=0xc000083f38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
runtime/mgc.go:1339 +0x25 fp=0xc000083fe0 sp=0xc000083fc8 pc=0x7ff792f84685
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000083fe8 sp=0xc000083fe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
runtime/mgc.go:1339 +0x105

goroutine 8 gp=0xc0003d8700 m=nil [GC worker (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000085f38 sp=0xc000085f18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008d960)
runtime/mgc.go:1423 +0xe9 fp=0xc000085fc8 sp=0xc000085f38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
runtime/mgc.go:1339 +0x25 fp=0xc000085fe0 sp=0xc000085fc8 pc=0x7ff792f84685
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000085fe8 sp=0xc000085fe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
runtime/mgc.go:1339 +0x105

goroutine 9 gp=0xc0003d88c0 m=nil [GC worker (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc00047bf38 sp=0xc00047bf18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008d960)
runtime/mgc.go:1423 +0xe9 fp=0xc00047bfc8 sp=0xc00047bf38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
runtime/mgc.go:1339 +0x25 fp=0xc00047bfe0 sp=0xc00047bfc8 pc=0x7ff792f84685
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc00047bfe8 sp=0xc00047bfe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
runtime/mgc.go:1339 +0x105

goroutine 18 gp=0xc0004861c0 m=nil [GC worker (idle)]:
runtime.gopark(0x4bd0e233d4d0?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000477f38 sp=0xc000477f18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008d960)
runtime/mgc.go:1423 +0xe9 fp=0xc000477fc8 sp=0xc000477f38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
runtime/mgc.go:1339 +0x25 fp=0xc000477fe0 sp=0xc000477fc8 pc=0x7ff792f84685
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000477fe8 sp=0xc000477fe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
runtime/mgc.go:1339 +0x105

goroutine 19 gp=0xc000486380 m=nil [GC worker (idle)]:
runtime.gopark(0x4bd0e233d4d0?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000479f38 sp=0xc000479f18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008d960)
runtime/mgc.go:1423 +0xe9 fp=0xc000479fc8 sp=0xc000479f38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
runtime/mgc.go:1339 +0x25 fp=0xc000479fe0 sp=0xc000479fc8 pc=0x7ff792f84685
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000479fe8 sp=0xc000479fe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
runtime/mgc.go:1339 +0x105

goroutine 20 gp=0xc000486540 m=nil [GC worker (idle)]:
runtime.gopark(0x4bd0e233d4d0?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000491f38 sp=0xc000491f18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008d960)
runtime/mgc.go:1423 +0xe9 fp=0xc000491fc8 sp=0xc000491f38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
runtime/mgc.go:1339 +0x25 fp=0xc000491fe0 sp=0xc000491fc8 pc=0x7ff792f84685
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000491fe8 sp=0xc000491fe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
runtime/mgc.go:1339 +0x105

goroutine 34 gp=0xc000206000 m=nil [GC worker (idle)]:
runtime.gopark(0x4bd0e233d4d0?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc00048df38 sp=0xc00048df18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008d960)
runtime/mgc.go:1423 +0xe9 fp=0xc00048dfc8 sp=0xc00048df38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
runtime/mgc.go:1339 +0x25 fp=0xc00048dfe0 sp=0xc00048dfc8 pc=0x7ff792f84685
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc00048dfe8 sp=0xc00048dfe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
runtime/mgc.go:1339 +0x105

goroutine 10 gp=0xc0003d8a80 m=nil [GC worker (idle)]:
runtime.gopark(0x4bd0e233d4d0?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc00047df38 sp=0xc00047df18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008d960)
runtime/mgc.go:1423 +0xe9 fp=0xc00047dfc8 sp=0xc00047df38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
runtime/mgc.go:1339 +0x25 fp=0xc00047dfe0 sp=0xc00047dfc8 pc=0x7ff792f84685
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc00047dfe8 sp=0xc00047dfe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
runtime/mgc.go:1339 +0x105

goroutine 11 gp=0xc0002061c0 m=nil [sync.WaitGroup.Wait]:
runtime.gopark(0x0?, 0x0?, 0x20?, 0x1?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc00048fe20 sp=0xc00048fe00 pc=0x7ff792fd598e
runtime.goparkunlock(...)
runtime/proc.go:441
runtime.semacquire1(0xc0003db100, 0x0, 0x1, 0x0, 0x18)
runtime/sema.go:188 +0x22f fp=0xc00048fe88 sp=0xc00048fe20 pc=0x7ff792fb750f
sync.runtime_SemacquireWaitGroup(0x0?)
runtime/sema.go:110 +0x25 fp=0xc00048fec0 sp=0xc00048fe88 pc=0x7ff792fd6f85
sync.(*WaitGroup).Wait(0x0?)
sync/waitgroup.go:118 +0x48 fp=0xc00048fee8 sp=0xc00048fec0 pc=0x7ff792feb988
github.com/ollama/ollama/runner/llamarunner.(*Server).run(0xc0003db0e0, {0x7ff7946e4a30, 0xc00033e550})
github.com/ollama/ollama/runner/llamarunner/runner.go:360 +0x4b fp=0xc00048ffb8 sp=0xc00048fee8 pc=0x7ff79352922b
github.com/ollama/ollama/runner/llamarunner.Execute.gowrap1()
github.com/ollama/ollama/runner/llamarunner/runner.go:981 +0x28 fp=0xc00048ffe0 sp=0xc00048ffb8 pc=0x7ff79352e788
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc00048ffe8 sp=0xc00048ffe0 pc=0x7ff792fdd9a1
created by github.com/ollama/ollama/runner/llamarunner.Execute in goroutine 1
github.com/ollama/ollama/runner/llamarunner/runner.go:981 +0x4c5

goroutine 12 gp=0xc000206380 m=nil [IO wait]:
runtime.gopark(0x0?, 0xc00047f420?, 0xc8?, 0xf4?, 0xc00047f4cc?)
runtime/proc.go:435 +0xce fp=0xc00004b8c8 sp=0xc00004b8a8 pc=0x7ff792fd598e
runtime.netpollblock(0x388?, 0x92f70406?, 0xf7?)
runtime/netpoll.go:575 +0xf7 fp=0xc00004b900 sp=0xc00004b8c8 pc=0x7ff792f9bdf7
internal/poll.runtime_pollWait(0x258f1615f38, 0x72)
runtime/netpoll.go:351 +0x85 fp=0xc00004b920 sp=0xc00004b900 pc=0x7ff792fd4b25
internal/poll.(*pollDesc).wait(0x7ff79319d5f7?, 0xc00004b970?, 0x0)
internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc00004b948 sp=0xc00004b920 pc=0x7ff79306bee7
internal/poll.execIO(0xc00047f420, 0x7ff794546098)
internal/poll/fd_windows.go:177 +0x105 fp=0xc00004b9c0 sp=0xc00004b948 pc=0x7ff79306d345
internal/poll.(*FD).Read(0xc00047f408, {0xc000091000, 0x1000, 0x1000})
internal/poll/fd_windows.go:438 +0x29b fp=0xc00004ba60 sp=0xc00004b9c0 pc=0x7ff79306e01b
net.(*netFD).Read(0xc00047f408, {0xc000091000?, 0xc00004bad0?, 0x7ff79306c3a5?})
net/fd_posix.go:55 +0x25 fp=0xc00004baa8 sp=0xc00004ba60 pc=0x7ff7930e1325
net.(*conn).Read(0xc0004ac530, {0xc000091000?, 0x0?, 0x0?})
net/net.go:194 +0x45 fp=0xc00004baf0 sp=0xc00004baa8 pc=0x7ff7930f0a45
net/http.(*connReader).Read(0xc0003242d0, {0xc000091000, 0x1000, 0x1000})
net/http/server.go:798 +0x159 fp=0xc00004bb40 sp=0xc00004baf0 pc=0x7ff7932de639
bufio.(*Reader).fill(0xc000200660)
bufio/bufio.go:113 +0x103 fp=0xc00004bb78 sp=0xc00004bb40 pc=0x7ff793107963
bufio.(*Reader).Peek(0xc000200660, 0x4)
bufio/bufio.go:152 +0x53 fp=0xc00004bb98 sp=0xc00004bb78 pc=0x7ff793107a93
net/http.(*conn).serve(0xc000204480, {0x7ff7946e49f8, 0xc000324000})
net/http/server.go:2137 +0x785 fp=0xc00004bfb8 sp=0xc00004bb98 pc=0x7ff7932e4425
net/http.(*Server).Serve.gowrap3()
net/http/server.go:3454 +0x28 fp=0xc00004bfe0 sp=0xc00004bfb8 pc=0x7ff7932e9b88
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc00004bfe8 sp=0xc00004bfe0 pc=0x7ff792fdd9a1
created by net/http.(*Server).Serve in goroutine 1
net/http/server.go:3454 +0x485
rax 0xffffffeb02005900
rbx 0x258fc0a4460
rcx 0x258f4258d40
rdx 0x0
rdi 0x0
rsi 0x258f42b3230
rbp 0xf997aff9c0
rsp 0xf997aff710
r8 0x258ac180f00
r9 0x1
r10 0x8000
r11 0xf997aff720
r12 0x667
r13 0x1218
r14 0x1
r15 0x258f4221790
rip 0x7ffceb1fe53e
rflags 0x10287
cs 0x33
fs 0x53
gs 0x2b
time=2026-02-17T16:20:43.383+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server not responding"
time=2026-02-17T16:20:43.634+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server error"
time=2026-02-17T16:20:44.386+01:00 level=INFO source=sched.go:493 msg="Load failed" model=F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e error="llama runner process has terminated: exit status 2"
[GIN] 2026/02/17 - 16:20:44 | 500 | 27.5168873s | 127.0.0.1 | POST "/api/chat"
[GIN] 2026/02/17 - 16:20:49 | 200 | 134.789ms | 127.0.0.1 | POST "/api/show"
[GIN] 2026/02/17 - 16:20:56 | 200 | 8.8411ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:21:00 | 200 | 11.267ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:21:00 | 200 | 135.358ms | 127.0.0.1 | POST "/api/show"
[GIN] 2026/02/17 - 16:21:01 | 200 | 119.4627ms | 127.0.0.1 | POST "/api/show"
time=2026-02-17T16:21:01.197+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\Users\robob\AppData\Local\Programs\Ollama\ollama.exe runner --ollama-engine --port 34095"
time=2026-02-17T16:21:01.673+01:00 level=INFO source=cpu_windows.go:148 msg=packages count=1
time=2026-02-17T16:21:01.673+01:00 level=INFO source=cpu_windows.go:195 msg="" package=0 cores=4 efficiency=0 threads=8
llama_model_loader: loaded meta data with 26 key-value pairs and 579 tensors from F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e (version GGUF V3 (latest))
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
llama_model_loader: - kv 0: general.architecture str = qwen2
llama_model_loader: - kv 1: general.type str = model
llama_model_loader: - kv 2: general.name str = DeepSeek R1 Distill Qwen 14B
llama_model_loader: - kv 3: general.basename str = DeepSeek-R1-Distill-Qwen
llama_model_loader: - kv 4: general.size_label str = 14B
llama_model_loader: - kv 5: qwen2.block_count u32 = 48
llama_model_loader: - kv 6: qwen2.context_length u32 = 131072
llama_model_loader: - kv 7: qwen2.embedding_length u32 = 5120
llama_model_loader: - kv 8: qwen2.feed_forward_length u32 = 13824
llama_model_loader: - kv 9: qwen2.attention.head_count u32 = 40
llama_model_loader: - kv 10: qwen2.attention.head_count_kv u32 = 8
llama_model_loader: - kv 11: qwen2.rope.freq_base f32 = 1000000.000000
llama_model_loader: - kv 12: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010
llama_model_loader: - kv 13: general.file_type u32 = 15
llama_model_loader: - kv 14: tokenizer.ggml.model str = gpt2
llama_model_loader: - kv 15: tokenizer.ggml.pre str = qwen2
llama_model_loader: - kv 16: tokenizer.ggml.tokens arr[str,152064] = ["!", """, "#", "$", "%", "&", "'", ...
llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv 18: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 151646
llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151643
llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643
llama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = true
llama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false
llama_model_loader: - kv 24: tokenizer.chat_template str = {% if not add_generation_prompt is de...
llama_model_loader: - kv 25: general.quantization_version u32 = 2
llama_model_loader: - type f32: 241 tensors
llama_model_loader: - type q4_K: 289 tensors
llama_model_loader: - type q6_K: 49 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type = Q4_K - Medium
print_info: file size = 8.37 GiB (4.87 BPW)
load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
load: printing all EOG tokens:
load: - 151643 ('<｜end▁of▁sentence｜>')
load: - 151662 ('<|fim_pad|>')
load: - 151663 ('<|repo_name|>')
load: - 151664 ('<|file_sep|>')
load: special tokens cache size = 22
load: token to piece cache size = 0.9310 MB
print_info: arch = qwen2
print_info: vocab_only = 1
print_info: no_alloc = 0
print_info: model type = ?B
print_info: model params = 14.77 B
print_info: general.name = DeepSeek R1 Distill Qwen 14B
print_info: vocab type = BPE
print_info: n_vocab = 152064
print_info: n_merges = 151387
print_info: BOS token = 151646 '<｜begin▁of▁sentence｜>'
print_info: EOS token = 151643 '<｜end▁of▁sentence｜>'
print_info: EOT token = 151643 '<｜end▁of▁sentence｜>'
print_info: PAD token = 151643 '<｜end▁of▁sentence｜>'
print_info: LF token = 198 'Ċ'
print_info: FIM PRE token = 151659 '<|fim_prefix|>'
print_info: FIM SUF token = 151661 '<|fim_suffix|>'
print_info: FIM MID token = 151660 '<|fim_middle|>'
print_info: FIM PAD token = 151662 '<|fim_pad|>'
print_info: FIM REP token = 151663 '<|repo_name|>'
print_info: FIM SEP token = 151664 '<|file_sep|>'
print_info: EOG token = 151643 '<｜end▁of▁sentence｜>'
print_info: EOG token = 151662 '<|fim_pad|>'
print_info: EOG token = 151663 '<|repo_name|>'
print_info: EOG token = 151664 '<|file_sep|>'
print_info: max token length = 256
llama_model_load: vocab only - skipping tensors
time=2026-02-17T16:21:02.012+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\Users\robob\AppData\Local\Programs\Ollama\ollama.exe runner --model F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e --port 34125"
time=2026-02-17T16:21:02.013+01:00 level=INFO source=sched.go:466 msg="system memory" total="15.9 GiB" free="9.3 GiB" free_swap="16.8 GiB"
time=2026-02-17T16:21:02.013+01:00 level=INFO source=sched.go:473 msg="gpu memory" id=GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 library=CUDA available="3.1 GiB" free="3.6 GiB" minimum="457.0 MiB" overhead="0 B"
time=2026-02-17T16:21:02.014+01:00 level=INFO source=server.go:498 msg="loading model" "model layers"=49 requested=1
time=2026-02-17T16:21:02.014+01:00 level=INFO source=device.go:240 msg="model weights" device=CUDA0 size="166.4 MiB"
time=2026-02-17T16:21:02.014+01:00 level=INFO source=device.go:245 msg="model weights" device=CPU size="7.8 GiB"
time=2026-02-17T16:21:02.014+01:00 level=INFO source=device.go:251 msg="kv cache" device=CUDA0 size="16.0 MiB"
time=2026-02-17T16:21:02.014+01:00 level=INFO source=device.go:256 msg="kv cache" device=CPU size="752.0 MiB"
time=2026-02-17T16:21:02.014+01:00 level=INFO source=device.go:262 msg="compute graph" device=CUDA0 size="916.1 MiB"
time=2026-02-17T16:21:02.014+01:00 level=INFO source=device.go:272 msg="total memory" size="9.6 GiB"
time=2026-02-17T16:21:02.058+01:00 level=INFO source=runner.go:965 msg="starting go runner"
load_backend: loaded CPU backend from C:\Users\robob\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll
ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 1 CUDA devices:
Device 0: NVIDIA GeForce GTX 980, compute capability 5.2, VMM: yes, ID: GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2
load_backend: loaded CUDA backend from C:\Users\robob\AppData\Local\Programs\Ollama\lib\ollama\cuda_v12\ggml-cuda.dll
time=2026-02-17T16:21:02.251+01:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 CUDA.0.ARCHS=500,520,600,610,700,750,800,860,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(clang)
time=2026-02-17T16:21:02.252+01:00 level=INFO source=runner.go:1001 msg="Server listening on 127.0.0.1:34125"
time=2026-02-17T16:21:02.255+01:00 level=INFO source=runner.go:895 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Auto KvSize:4096 KvCacheType: NumThreads:4 GPULayers:1[ID:GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 Layers:1(47..47)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2026-02-17T16:21:02.255+01:00 level=INFO source=server.go:1350 msg="waiting for llama runner to start responding"
time=2026-02-17T16:21:02.255+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server loading model"
ggml_backend_cuda_device_get_memory device GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 utilizing NVML memory reporting free: 3815043072 total: 4294967296
llama_model_load_from_file_impl: using device CUDA0 (NVIDIA GeForce GTX 980) (0000:01:00.0) - 3638 MiB free
llama_model_loader: loaded meta data with 26 key-value pairs and 579 tensors from F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e (version GGUF V3 (latest))
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
llama_model_loader: - kv 0: general.architecture str = qwen2
llama_model_loader: - kv 1: general.type str = model
llama_model_loader: - kv 2: general.name str = DeepSeek R1 Distill Qwen 14B
llama_model_loader: - kv 3: general.basename str = DeepSeek-R1-Distill-Qwen
llama_model_loader: - kv 4: general.size_label str = 14B
llama_model_loader: - kv 5: qwen2.block_count u32 = 48
llama_model_loader: - kv 6: qwen2.context_length u32 = 131072
llama_model_loader: - kv 7: qwen2.embedding_length u32 = 5120
llama_model_loader: - kv 8: qwen2.feed_forward_length u32 = 13824
llama_model_loader: - kv 9: qwen2.attention.head_count u32 = 40
llama_model_loader: - kv 10: qwen2.attention.head_count_kv u32 = 8
llama_model_loader: - kv 11: qwen2.rope.freq_base f32 = 1000000.000000
llama_model_loader: - kv 12: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010
llama_model_loader: - kv 13: general.file_type u32 = 15
llama_model_loader: - kv 14: tokenizer.ggml.model str = gpt2
llama_model_loader: - kv 15: tokenizer.ggml.pre str = qwen2
llama_model_loader: - kv 16: tokenizer.ggml.tokens arr[str,152064] = ["!", """, "#", "$", "%", "&", "'", ...
llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv 18: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 151646
llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151643
llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643
llama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = true
llama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false
llama_model_loader: - kv 24: tokenizer.chat_template str = {% if not add_generation_prompt is de...
llama_model_loader: - kv 25: general.quantization_version u32 = 2
llama_model_loader: - type f32: 241 tensors
llama_model_loader: - type q4_K: 289 tensors
llama_model_loader: - type q6_K: 49 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type = Q4_K - Medium
print_info: file size = 8.37 GiB (4.87 BPW)
load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
load: printing all EOG tokens:
load: - 151643 ('<｜end▁of▁sentence｜>')
load: - 151662 ('<|fim_pad|>')
load: - 151663 ('<|repo_name|>')
load: - 151664 ('<|file_sep|>')
load: special tokens cache size = 22
load: token to piece cache size = 0.9310 MB
print_info: arch = qwen2
print_info: vocab_only = 0
print_info: no_alloc = 0
print_info: n_ctx_train = 131072
print_info: n_embd = 5120
print_info: n_embd_inp = 5120
print_info: n_layer = 48
print_info: n_head = 40
print_info: n_head_kv = 8
print_info: n_rot = 128
print_info: n_swa = 0
print_info: is_swa_any = 0
print_info: n_embd_head_k = 128
print_info: n_embd_head_v = 128
print_info: n_gqa = 5
print_info: n_embd_k_gqa = 1024
print_info: n_embd_v_gqa = 1024
print_info: f_norm_eps = 0.0e+00
print_info: f_norm_rms_eps = 1.0e-05
print_info: f_clamp_kqv = 0.0e+00
print_info: f_max_alibi_bias = 0.0e+00
print_info: f_logit_scale = 0.0e+00
print_info: f_attn_scale = 0.0e+00
print_info: n_ff = 13824
print_info: n_expert = 0
print_info: n_expert_used = 0
print_info: n_expert_groups = 0
print_info: n_group_used = 0
print_info: causal attn = 1
print_info: pooling type = -1
print_info: rope type = 2
print_info: rope scaling = linear
print_info: freq_base_train = 1000000.0
print_info: freq_scale_train = 1
print_info: n_ctx_orig_yarn = 131072
print_info: rope_yarn_log_mul= 0.0000
print_info: rope_finetuned = unknown
print_info: model type = 14B
print_info: model params = 14.77 B
print_info: general.name = DeepSeek R1 Distill Qwen 14B
print_info: vocab type = BPE
print_info: n_vocab = 152064
print_info: n_merges = 151387
print_info: BOS token = 151646 '<｜begin▁of▁sentence｜>'
print_info: EOS token = 151643 '<｜end▁of▁sentence｜>'
print_info: EOT token = 151643 '<｜end▁of▁sentence｜>'
print_info: PAD token = 151643 '<｜end▁of▁sentence｜>'
print_info: LF token = 198 'Ċ'
print_info: FIM PRE token = 151659 '<|fim_prefix|>'
print_info: FIM SUF token = 151661 '<|fim_suffix|>'
print_info: FIM MID token = 151660 '<|fim_middle|>'
print_info: FIM PAD token = 151662 '<|fim_pad|>'
print_info: FIM REP token = 151663 '<|repo_name|>'
print_info: FIM SEP token = 151664 '<|file_sep|>'
print_info: EOG token = 151643 '<｜end▁of▁sentence｜>'
print_info: EOG token = 151662 '<|fim_pad|>'
print_info: EOG token = 151663 '<|repo_name|>'
print_info: EOG token = 151664 '<|file_sep|>'
print_info: max token length = 256
load_tensors: loading model tensors, this can take a while... (mmap = false)
ggml_cuda_host_malloc: failed to allocate 7981.97 MiB of pinned memory: out of memory
load_tensors: offloading 1 repeating layers to GPU
load_tensors: offloaded 1/49 layers to GPU
load_tensors: CPU model buffer size = 417.66 MiB
load_tensors: CUDA0 model buffer size = 166.41 MiB
load_tensors: CPU model buffer size = 7981.97 MiB
llama_context: constructing llama_context
llama_context: n_seq_max = 1
llama_context: n_ctx = 4096
llama_context: n_ctx_seq = 4096
llama_context: n_batch = 512
llama_context: n_ubatch = 512
llama_context: causal_attn = 1
llama_context: flash_attn = auto
llama_context: kv_unified = false
llama_context: freq_base = 1000000.0
llama_context: freq_scale = 1
llama_context: n_ctx_seq (4096) < n_ctx_train (131072) -- the full capacity of the model will not be utilized
llama_context: CPU output buffer size = 0.60 MiB
llama_kv_cache: CPU KV buffer size = 752.00 MiB
llama_kv_cache: CUDA0 KV buffer size = 16.00 MiB
llama_kv_cache: size = 768.00 MiB ( 4096 cells, 48 layers, 1/1 seqs), K (f16): 384.00 MiB, V (f16): 384.00 MiB
llama_context: Flash Attention was auto, set to enabled
graph_reserve: failed to allocate compute buffers
Exception 0xc0000005 0x0 0x1d7cc1fea10 0x7ffceb1fe53e
PC=0x7ffceb1fe53e
signal arrived during external code execution

runtime.cgocall(0x7ff793e6e2b0, 0xc00022fc00)
runtime/cgocall.go:167 +0x3e fp=0xc00022fbd8 sp=0xc00022fb70 pc=0x7ff792fd243e
github.com/ollama/ollama/llama._Cfunc_llama_init_from_model(0x1eafd0e00b0, {0x1000, 0x200, 0x200, 0x1, 0x4, 0x4, 0xffffffff, 0xffffffff, 0xffffffff, ...})
_cgo_gotypes.go:759 +0x54 fp=0xc00022fc00 sp=0xc00022fbd8 pc=0x7ff793479774
github.com/ollama/ollama/llama.NewContextWithModel.func1(...)
github.com/ollama/ollama/llama/llama.go:322
github.com/ollama/ollama/llama.NewContextWithModel(0xc0001cd1e0, {{0x1000, 0x200, 0x200, 0x1, 0x4, 0x4, 0xffffffff, 0xffffffff, 0xffffffff, ...}})
github.com/ollama/ollama/llama/llama.go:322 +0x158 fp=0xc00022fda0 sp=0xc00022fc00 pc=0x7ff79347dd58
github.com/ollama/ollama/runner/llamarunner.(*Server).loadModel(0xc00043fae0, {{0xc0003806f0, 0x1, 0x1}, 0x1, 0x0, 0x0, {0xc0003806e8, 0x1, 0x2}, ...}, ...)
github.com/ollama/ollama/runner/llamarunner/runner.go:847 +0x178 fp=0xc00022fee8 sp=0xc00022fda0 pc=0x7ff79352c938
github.com/ollama/ollama/runner/llamarunner.(*Server).load.gowrap2()
github.com/ollama/ollama/runner/llamarunner/runner.go:934 +0x114 fp=0xc00022ffe0 sp=0xc00022fee8 pc=0x7ff79352db54
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc00022ffe8 sp=0xc00022ffe0 pc=0x7ff792fdd9a1
created by github.com/ollama/ollama/runner/llamarunner.(*Server).load in goroutine 11
github.com/ollama/ollama/runner/llamarunner/runner.go:934 +0x889

goroutine 1 gp=0xc0000021c0 m=nil [IO wait]:
runtime.gopark(0x7ff792fdf1a0?, 0x7ff795180d00?, 0x20?, 0xd4?, 0xc00060d4cc?)
runtime/proc.go:435 +0xce fp=0xc0004c9630 sp=0xc0004c9610 pc=0x7ff792fd598e
runtime.netpollblock(0x384?, 0x92f70406?, 0xf7?)
runtime/netpoll.go:575 +0xf7 fp=0xc0004c9668 sp=0xc0004c9630 pc=0x7ff792f9bdf7
internal/poll.runtime_pollWait(0x1eafa29c470, 0x72)
runtime/netpoll.go:351 +0x85 fp=0xc0004c9688 sp=0xc0004c9668 pc=0x7ff792fd4b25
internal/poll.(*pollDesc).wait(0x7ff79306a8f3?, 0x0?, 0x0)
internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc0004c96b0 sp=0xc0004c9688 pc=0x7ff79306bee7
internal/poll.execIO(0xc00060d420, 0xc00048d758)
internal/poll/fd_windows.go:177 +0x105 fp=0xc0004c9728 sp=0xc0004c96b0 pc=0x7ff79306d345
internal/poll.(*FD).acceptOne(0xc00060d408, 0x390, {0xc0006500f0?, 0xc00048d7b8?, 0x7ff793075005?}, 0xc00048d7ec?)
internal/poll/fd_windows.go:946 +0x65 fp=0xc0004c9788 sp=0xc0004c9728 pc=0x7ff7930718c5
internal/poll.(*FD).Accept(0xc00060d408, 0xc0004c9938)
internal/poll/fd_windows.go:980 +0x1b6 fp=0xc0004c9840 sp=0xc0004c9788 pc=0x7ff793071bf6
net.(*netFD).accept(0xc00060d408)
net/fd_windows.go:182 +0x4b fp=0xc0004c9958 sp=0xc0004c9840 pc=0x7ff7930e344b
net.(*TCPListener).accept(0xc0003d46c0)
net/tcpsock_posix.go:159 +0x1b fp=0xc0004c99a8 sp=0xc0004c9958 pc=0x7ff7930f99fb
net.(*TCPListener).Accept(0xc0003d46c0)
net/tcpsock.go:380 +0x30 fp=0xc0004c99d8 sp=0xc0004c99a8 pc=0x7ff7930f87b0
net/http.(*onceCloseListener).Accept(0xc000632480?)
:1 +0x24 fp=0xc0004c99f0 sp=0xc0004c99d8 pc=0x7ff793311ec4
net/http.(*Server).Serve(0xc000141d00, {0x7ff7946e21d0, 0xc0003d46c0})
net/http/server.go:3424 +0x30c fp=0xc0004c9b20 sp=0xc0004c99f0 pc=0x7ff7932e978c
github.com/ollama/ollama/runner/llamarunner.Execute({0xc0000a4020, 0x4, 0x6})
github.com/ollama/ollama/runner/llamarunner/runner.go:1002 +0x8f5 fp=0xc0004c9cf0 sp=0xc0004c9b20 pc=0x7ff79352e515
github.com/ollama/ollama/runner.Execute({0xc0000a4010?, 0x0?, 0x0?})
github.com/ollama/ollama/runner/runner.go:25 +0x1a5 fp=0xc0004c9d30 sp=0xc0004c9cf0 pc=0x7ff793603ae5
github.com/ollama/ollama/cmd.NewCLI.func3(0xc000141800?, {0x7ff7944c4587?, 0x4?, 0x7ff7944c458b?})
github.com/ollama/ollama/cmd/cmd.go:2262 +0x45 fp=0xc0004c9d58 sp=0xc0004c9d30 pc=0x7ff793dfce65
github.com/spf13/cobra.(*Command).execute(0xc000149508, {0xc0003d4440, 0x4, 0x4})
github.com/spf13/cobra@v1.7.0/command.go:940 +0x85c fp=0xc0004c9e78 sp=0xc0004c9d58 pc=0x7ff79315e61c
github.com/spf13/cobra.(*Command).ExecuteC(0xc000714c08)
github.com/spf13/cobra@v1.7.0/command.go:1068 +0x3a5 fp=0xc0004c9f30 sp=0xc0004c9e78 pc=0x7ff79315ee65
github.com/spf13/cobra.(*Command).Execute(...)
github.com/spf13/cobra@v1.7.0/command.go:992
github.com/spf13/cobra.(*Command).ExecuteContext(...)
github.com/spf13/cobra@v1.7.0/command.go:985
main.main()
github.com/ollama/ollama/main.go:12 +0x4d fp=0xc0004c9f50 sp=0xc0004c9f30 pc=0x7ff793dff28d
runtime.main()
runtime/proc.go:283 +0x27d fp=0xc0004c9fe0 sp=0xc0004c9f50 pc=0x7ff792fa4ddd
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc0004c9fe8 sp=0xc0004c9fe0 pc=0x7ff792fdd9a1

goroutine 2 gp=0xc0000028c0 m=nil [force gc (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000071fa8 sp=0xc000071f88 pc=0x7ff792fd598e
runtime.goparkunlock(...)
runtime/proc.go:441
runtime.forcegchelper()
runtime/proc.go:348 +0xb8 fp=0xc000071fe0 sp=0xc000071fa8 pc=0x7ff792fa50f8
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000071fe8 sp=0xc000071fe0 pc=0x7ff792fdd9a1
created by runtime.init.7 in goroutine 1
runtime/proc.go:336 +0x1a

goroutine 3 gp=0xc000002c40 m=nil [GC sweep wait]:
runtime.gopark(0x1?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000073f80 sp=0xc000073f60 pc=0x7ff792fd598e
runtime.goparkunlock(...)
runtime/proc.go:441
runtime.bgsweep(0xc000080000)
runtime/mgcsweep.go:316 +0xdf fp=0xc000073fc8 sp=0xc000073f80 pc=0x7ff792f8debf
runtime.gcenable.gowrap1()
runtime/mgc.go:204 +0x25 fp=0xc000073fe0 sp=0xc000073fc8 pc=0x7ff792f82285
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000073fe8 sp=0xc000073fe0 pc=0x7ff792fdd9a1
created by runtime.gcenable in goroutine 1
runtime/mgc.go:204 +0x66

goroutine 4 gp=0xc000002e00 m=nil [GC scavenge wait]:
runtime.gopark(0x10000?, 0x7ff7946ccc38?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000087f78 sp=0xc000087f58 pc=0x7ff792fd598e
runtime.goparkunlock(...)
runtime/proc.go:441
runtime.(*scavengerState).park(0x7ff7951aa900)
runtime/mgcscavenge.go:425 +0x49 fp=0xc000087fa8 sp=0xc000087f78 pc=0x7ff792f8b909
runtime.bgscavenge(0xc000080000)
runtime/mgcscavenge.go:658 +0x59 fp=0xc000087fc8 sp=0xc000087fa8 pc=0x7ff792f8be99
runtime.gcenable.gowrap2()
runtime/mgc.go:205 +0x25 fp=0xc000087fe0 sp=0xc000087fc8 pc=0x7ff792f82225
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000087fe8 sp=0xc000087fe0 pc=0x7ff792fdd9a1
created by runtime.gcenable in goroutine 1
runtime/mgc.go:205 +0xa5

goroutine 5 gp=0xc000003340 m=nil [finalizer wait]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000089e30 sp=0xc000089e10 pc=0x7ff792fd598e
runtime.runfinq()
runtime/mfinal.go:196 +0x107 fp=0xc000089fe0 sp=0xc000089e30 pc=0x7ff792f81207
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000089fe8 sp=0xc000089fe0 pc=0x7ff792fdd9a1
created by runtime.createfing in goroutine 1
runtime/mfinal.go:166 +0x3d

goroutine 6 gp=0xc000003dc0 m=nil [chan receive]:
runtime.gopark(0xc0000d5a40?, 0xc000510300?, 0x60?, 0x5f?, 0x7ff7930cc148?)
runtime/proc.go:435 +0xce fp=0xc000075f18 sp=0xc000075ef8 pc=0x7ff792fd598e
runtime.chanrecv(0xc00008c380, 0x0, 0x1)
runtime/chan.go:664 +0x445 fp=0xc000075f90 sp=0xc000075f18 pc=0x7ff792f72d45
runtime.chanrecv1(0x7ff792fa4f40?, 0xc000075f76?)
runtime/chan.go:506 +0x12 fp=0xc000075fb8 sp=0xc000075f90 pc=0x7ff792f728d2
runtime.unique_runtime_registerUniqueMapCleanup.func2(...)
runtime/mgc.go:1796
runtime.unique_runtime_registerUniqueMapCleanup.gowrap1()
runtime/mgc.go:1799 +0x2f fp=0xc000075fe0 sp=0xc000075fb8 pc=0x7ff792f854af
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000075fe8 sp=0xc000075fe0 pc=0x7ff792fdd9a1
created by unique.runtime_registerUniqueMapCleanup in goroutine 1
runtime/mgc.go:1794 +0x85

goroutine 7 gp=0xc0003da540 m=nil [GC worker (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000083f38 sp=0xc000083f18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008d960)
runtime/mgc.go:1423 +0xe9 fp=0xc000083fc8 sp=0xc000083f38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
runtime/mgc.go:1339 +0x25 fp=0xc000083fe0 sp=0xc000083fc8 pc=0x7ff792f84685
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000083fe8 sp=0xc000083fe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
runtime/mgc.go:1339 +0x105

goroutine 8 gp=0xc0003da700 m=nil [GC worker (idle)]:
runtime.gopark(0x4bde8a0df494?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000085f38 sp=0xc000085f18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008d960)
runtime/mgc.go:1423 +0xe9 fp=0xc000085fc8 sp=0xc000085f38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
runtime/mgc.go:1339 +0x25 fp=0xc000085fe0 sp=0xc000085fc8 pc=0x7ff792f84685
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000085fe8 sp=0xc000085fe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
runtime/mgc.go:1339 +0x105

goroutine 9 gp=0xc0003da8c0 m=nil [GC worker (idle)]:
runtime.gopark(0x4bde8a0df494?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc00047bf38 sp=0xc00047bf18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008d960)
runtime/mgc.go:1423 +0xe9 fp=0xc00047bfc8 sp=0xc00047bf38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
runtime/mgc.go:1339 +0x25 fp=0xc00047bfe0 sp=0xc00047bfc8 pc=0x7ff792f84685
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc00047bfe8 sp=0xc00047bfe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
runtime/mgc.go:1339 +0x105

goroutine 18 gp=0xc0002081c0 m=nil [GC worker (idle)]:
runtime.gopark(0x4bde8a0df494?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000477f38 sp=0xc000477f18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008d960)
runtime/mgc.go:1423 +0xe9 fp=0xc000477fc8 sp=0xc000477f38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
runtime/mgc.go:1339 +0x25 fp=0xc000477fe0 sp=0xc000477fc8 pc=0x7ff792f84685
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000477fe8 sp=0xc000477fe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
runtime/mgc.go:1339 +0x105

goroutine 19 gp=0xc000208380 m=nil [GC worker (idle)]:
runtime.gopark(0x4bde8a0df494?, 0x3?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000479f38 sp=0xc000479f18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008d960)
runtime/mgc.go:1423 +0xe9 fp=0xc000479fc8 sp=0xc000479f38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
runtime/mgc.go:1339 +0x25 fp=0xc000479fe0 sp=0xc000479fc8 pc=0x7ff792f84685
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000479fe8 sp=0xc000479fe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
runtime/mgc.go:1339 +0x105

goroutine 20 gp=0xc000208540 m=nil [GC worker (idle)]:
runtime.gopark(0x4bde8a0df494?, 0x0?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000213f38 sp=0xc000213f18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008d960)
runtime/mgc.go:1423 +0xe9 fp=0xc000213fc8 sp=0xc000213f38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
runtime/mgc.go:1339 +0x25 fp=0xc000213fe0 sp=0xc000213fc8 pc=0x7ff792f84685
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000213fe8 sp=0xc000213fe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
runtime/mgc.go:1339 +0x105

goroutine 34 gp=0xc000484000 m=nil [GC worker (idle)]:
runtime.gopark(0x4bde8a0df494?, 0x3?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc00020ff38 sp=0xc00020ff18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008d960)
runtime/mgc.go:1423 +0xe9 fp=0xc00020ffc8 sp=0xc00020ff38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
runtime/mgc.go:1339 +0x25 fp=0xc00020ffe0 sp=0xc00020ffc8 pc=0x7ff792f84685
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc00020ffe8 sp=0xc00020ffe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
runtime/mgc.go:1339 +0x105

goroutine 21 gp=0xc000208700 m=nil [GC worker (idle)]:
runtime.gopark(0x7ff7951fc600?, 0x1?, 0x0?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000215f38 sp=0xc000215f18 pc=0x7ff792fd598e
runtime.gcBgMarkWorker(0xc00008d960)
runtime/mgc.go:1423 +0xe9 fp=0xc000215fc8 sp=0xc000215f38 pc=0x7ff792f847a9
runtime.gcBgMarkStartWorkers.gowrap1()
runtime/mgc.go:1339 +0x25 fp=0xc000215fe0 sp=0xc000215fc8 pc=0x7ff792f84685
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000215fe8 sp=0xc000215fe0 pc=0x7ff792fdd9a1
created by runtime.gcBgMarkStartWorkers in goroutine 1
runtime/mgc.go:1339 +0x105

goroutine 10 gp=0xc000484a80 m=nil [sync.WaitGroup.Wait]:
runtime.gopark(0x0?, 0x0?, 0x60?, 0x0?, 0x0?)
runtime/proc.go:435 +0xce fp=0xc000211e20 sp=0xc000211e00 pc=0x7ff792fd598e
runtime.goparkunlock(...)
runtime/proc.go:441
runtime.semacquire1(0xc00043fb00, 0x0, 0x1, 0x0, 0x18)
runtime/sema.go:188 +0x22f fp=0xc000211e88 sp=0xc000211e20 pc=0x7ff792fb750f
sync.runtime_SemacquireWaitGroup(0x0?)
runtime/sema.go:110 +0x25 fp=0xc000211ec0 sp=0xc000211e88 pc=0x7ff792fd6f85
sync.(*WaitGroup).Wait(0x0?)
sync/waitgroup.go:118 +0x48 fp=0xc000211ee8 sp=0xc000211ec0 pc=0x7ff792feb988
github.com/ollama/ollama/runner/llamarunner.(*Server).run(0xc00043fae0, {0x7ff7946e4a30, 0xc00033e2d0})
github.com/ollama/ollama/runner/llamarunner/runner.go:360 +0x4b fp=0xc000211fb8 sp=0xc000211ee8 pc=0x7ff79352922b
github.com/ollama/ollama/runner/llamarunner.Execute.gowrap1()
github.com/ollama/ollama/runner/llamarunner/runner.go:981 +0x28 fp=0xc000211fe0 sp=0xc000211fb8 pc=0x7ff79352e788
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc000211fe8 sp=0xc000211fe0 pc=0x7ff792fdd9a1
created by github.com/ollama/ollama/runner/llamarunner.Execute in goroutine 1
github.com/ollama/ollama/runner/llamarunner/runner.go:981 +0x4c5

goroutine 11 gp=0xc000484c40 m=nil [IO wait]:
runtime.gopark(0x0?, 0xc00060d6a0?, 0x48?, 0xd7?, 0xc00060d74c?)
runtime/proc.go:435 +0xce fp=0xc0004c58c8 sp=0xc0004c58a8 pc=0x7ff792fd598e
runtime.netpollblock(0x38c?, 0x92f70406?, 0xf7?)
runtime/netpoll.go:575 +0xf7 fp=0xc0004c5900 sp=0xc0004c58c8 pc=0x7ff792f9bdf7
internal/poll.runtime_pollWait(0x1eafa29c358, 0x72)
runtime/netpoll.go:351 +0x85 fp=0xc0004c5920 sp=0xc0004c5900 pc=0x7ff792fd4b25
internal/poll.(*pollDesc).wait(0x7ff79319d5f7?, 0xc0004c5970?, 0x0)
internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc0004c5948 sp=0xc0004c5920 pc=0x7ff79306bee7
internal/poll.execIO(0xc00060d6a0, 0x7ff794546098)
internal/poll/fd_windows.go:177 +0x105 fp=0xc0004c59c0 sp=0xc0004c5948 pc=0x7ff79306d345
internal/poll.(*FD).Read(0xc00060d688, {0xc000091000, 0x1000, 0x1000})
internal/poll/fd_windows.go:438 +0x29b fp=0xc0004c5a60 sp=0xc0004c59c0 pc=0x7ff79306e01b
net.(*netFD).Read(0xc00060d688, {0xc000091000?, 0xc0004c5ad0?, 0x7ff79306c3a5?})
net/fd_posix.go:55 +0x25 fp=0xc0004c5aa8 sp=0xc0004c5a60 pc=0x7ff7930e1325
net.(*conn).Read(0xc000078890, {0xc000091000?, 0x0?, 0x0?})
net/net.go:194 +0x45 fp=0xc0004c5af0 sp=0xc0004c5aa8 pc=0x7ff7930f0a45
net/http.(*connReader).Read(0xc000326090, {0xc000091000, 0x1000, 0x1000})
net/http/server.go:798 +0x159 fp=0xc0004c5b40 sp=0xc0004c5af0 pc=0x7ff7932de639
bufio.(*Reader).fill(0xc000193e60)
bufio/bufio.go:113 +0x103 fp=0xc0004c5b78 sp=0xc0004c5b40 pc=0x7ff793107963
bufio.(*Reader).Peek(0xc000193e60, 0x4)
bufio/bufio.go:152 +0x53 fp=0xc0004c5b98 sp=0xc0004c5b78 pc=0x7ff793107a93
net/http.(*conn).serve(0xc000632480, {0x7ff7946e49f8, 0xc0001c6090})
net/http/server.go:2137 +0x785 fp=0xc0004c5fb8 sp=0xc0004c5b98 pc=0x7ff7932e4425
net/http.(*Server).Serve.gowrap3()
net/http/server.go:3454 +0x28 fp=0xc0004c5fe0 sp=0xc0004c5fb8 pc=0x7ff7932e9b88
runtime.goexit({})
runtime/asm_amd64.s:1700 +0x1 fp=0xc0004c5fe8 sp=0xc0004c5fe0 pc=0x7ff792fdd9a1
created by net/http.(*Server).Serve in goroutine 1
net/http/server.go:3454 +0x485
rax 0xffffffed2d005900
rbx 0x1ea9f1f9110
rcx 0x1eafd2dbf10
rdx 0x0
rdi 0x0
rsi 0x1ea9f293a80
rbp 0x41d58ff930
rsp 0x41d58ff680
r8 0x1eab4d10fc0
r9 0x1
r10 0x8000
r11 0x41d58ff690
r12 0x667
r13 0x1218
r14 0x1
r15 0x1eafd0e00b0
rip 0x7ffceb1fe53e
rflags 0x10287
cs 0x33
fs 0x53
gs 0x2b
time=2026-02-17T16:21:25.001+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server error"
time=2026-02-17T16:21:26.003+01:00 level=INFO source=sched.go:493 msg="Load failed" model=F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e error="llama runner process has terminated: exit status 2"
[GIN] 2026/02/17 - 16:21:26 | 500 | 24.9481468s | 127.0.0.1 | POST "/api/chat"
[GIN] 2026/02/17 - 16:21:30 | 200 | 13.9638ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:21:37 | 200 | 127.8505ms | 127.0.0.1 | POST "/api/show"
[GIN] 2026/02/17 - 16:21:40 | 200 | 10.4352ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:21:42 | 200 | 9.3882ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2026/02/17 - 16:21:42 | 200 | 126.1458ms | 127.0.0.1 | POST "/api/show"
[GIN] 2026/02/17 - 16:21:42 | 200 | 118.4192ms | 127.0.0.1 | POST "/api/show"
time=2026-02-17T16:21:42.956+01:00 level=INFO source=cpu_windows.go:148 msg=packages count=1
time=2026-02-17T16:21:42.957+01:00 level=INFO source=cpu_windows.go:195 msg="" package=0 cores=4 efficiency=0 threads=8
llama_model_loader: loaded meta data with 26 key-value pairs and 579 tensors from F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e (version GGUF V3 (latest))
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
llama_model_loader: - kv 0: general.architecture str = qwen2
llama_model_loader: - kv 1: general.type str = model
llama_model_loader: - kv 2: general.name str = DeepSeek R1 Distill Qwen 14B
llama_model_loader: - kv 3: general.basename str = DeepSeek-R1-Distill-Qwen
llama_model_loader: - kv 4: general.size_label str = 14B
llama_model_loader: - kv 5: qwen2.block_count u32 = 48
llama_model_loader: - kv 6: qwen2.context_length u32 = 131072
llama_model_loader: - kv 7: qwen2.embedding_length u32 = 5120
llama_model_loader: - kv 8: qwen2.feed_forward_length u32 = 13824
llama_model_loader: - kv 9: qwen2.attention.head_count u32 = 40
llama_model_loader: - kv 10: qwen2.attention.head_count_kv u32 = 8
llama_model_loader: - kv 11: qwen2.rope.freq_base f32 = 1000000.000000
llama_model_loader: - kv 12: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010
llama_model_loader: - kv 13: general.file_type u32 = 15
llama_model_loader: - kv 14: tokenizer.ggml.model str = gpt2
llama_model_loader: - kv 15: tokenizer.ggml.pre str = qwen2
llama_model_loader: - kv 16: tokenizer.ggml.tokens arr[str,152064] = ["!", """, "#", "$", "%", "&", "'", ...
llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv 18: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 151646
llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151643
llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643
llama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = true
llama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false
llama_model_loader: - kv 24: tokenizer.chat_template str = {% if not add_generation_prompt is de...
llama_model_loader: - kv 25: general.quantization_version u32 = 2
llama_model_loader: - type f32: 241 tensors
llama_model_loader: - type q4_K: 289 tensors
llama_model_loader: - type q6_K: 49 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type = Q4_K - Medium
print_info: file size = 8.37 GiB (4.87 BPW)
load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
load: printing all EOG tokens:
load: - 151643 ('<｜end▁of▁sentence｜>')
load: - 151662 ('<|fim_pad|>')
load: - 151663 ('<|repo_name|>')
load: - 151664 ('<|file_sep|>')
load: special tokens cache size = 22
load: token to piece cache size = 0.9310 MB
print_info: arch = qwen2
print_info: vocab_only = 1
print_info: no_alloc = 0
print_info: model type = ?B
print_info: model params = 14.77 B
print_info: general.name = DeepSeek R1 Distill Qwen 14B
print_info: vocab type = BPE
print_info: n_vocab = 152064
print_info: n_merges = 151387
print_info: BOS token = 151646 '<｜begin▁of▁sentence｜>'
print_info: EOS token = 151643 '<｜end▁of▁sentence｜>'
print_info: EOT token = 151643 '<｜end▁of▁sentence｜>'
print_info: PAD token = 151643 '<｜end▁of▁sentence｜>'
print_info: LF token = 198 'Ċ'
print_info: FIM PRE token = 151659 '<|fim_prefix|>'
print_info: FIM SUF token = 151661 '<|fim_suffix|>'
print_info: FIM MID token = 151660 '<|fim_middle|>'
print_info: FIM PAD token = 151662 '<|fim_pad|>'
print_info: FIM REP token = 151663 '<|repo_name|>'
print_info: FIM SEP token = 151664 '<|file_sep|>'
print_info: EOG token = 151643 '<｜end▁of▁sentence｜>'
print_info: EOG token = 151662 '<|fim_pad|>'
print_info: EOG token = 151663 '<|repo_name|>'
print_info: EOG token = 151664 '<|file_sep|>'
print_info: max token length = 256
llama_model_load: vocab only - skipping tensors
time=2026-02-17T16:21:43.281+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\Users\robob\AppData\Local\Programs\Ollama\ollama.exe runner --model F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e --port 34156"
time=2026-02-17T16:21:43.282+01:00 level=INFO source=sched.go:466 msg="system memory" total="15.9 GiB" free="9.7 GiB" free_swap="17.0 GiB"
time=2026-02-17T16:21:43.282+01:00 level=INFO source=server.go:498 msg="loading model" "model layers"=49 requested=0
time=2026-02-17T16:21:43.283+01:00 level=INFO source=device.go:245 msg="model weights" device=CPU size="8.0 GiB"
time=2026-02-17T16:21:43.283+01:00 level=INFO source=device.go:256 msg="kv cache" device=CPU size="768.0 MiB"
time=2026-02-17T16:21:43.283+01:00 level=INFO source=device.go:272 msg="total memory" size="8.7 GiB"
time=2026-02-17T16:21:43.374+01:00 level=INFO source=runner.go:965 msg="starting go runner"
load_backend: loaded CPU backend from C:\Users\robob\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll
time=2026-02-17T16:21:43.443+01:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 compiler=cgo(clang)
time=2026-02-17T16:21:43.444+01:00 level=INFO source=runner.go:1001 msg="Server listening on 127.0.0.1:34156"
time=2026-02-17T16:21:43.445+01:00 level=INFO source=runner.go:895 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Auto KvSize:4096 KvCacheType: NumThreads:4 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2026-02-17T16:21:43.446+01:00 level=INFO source=server.go:1350 msg="waiting for llama runner to start responding"
time=2026-02-17T16:21:43.446+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server loading model"
llama_model_loader: loaded meta data with 26 key-value pairs and 579 tensors from F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e (version GGUF V3 (latest))
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
llama_model_loader: - kv 0: general.architecture str = qwen2
llama_model_loader: - kv 1: general.type str = model
llama_model_loader: - kv 2: general.name str = DeepSeek R1 Distill Qwen 14B
llama_model_loader: - kv 3: general.basename str = DeepSeek-R1-Distill-Qwen
llama_model_loader: - kv 4: general.size_label str = 14B
llama_model_loader: - kv 5: qwen2.block_count u32 = 48
llama_model_loader: - kv 6: qwen2.context_length u32 = 131072
llama_model_loader: - kv 7: qwen2.embedding_length u32 = 5120
llama_model_loader: - kv 8: qwen2.feed_forward_length u32 = 13824
llama_model_loader: - kv 9: qwen2.attention.head_count u32 = 40
llama_model_loader: - kv 10: qwen2.attention.head_count_kv u32 = 8
llama_model_loader: - kv 11: qwen2.rope.freq_base f32 = 1000000.000000
llama_model_loader: - kv 12: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010
llama_model_loader: - kv 13: general.file_type u32 = 15
llama_model_loader: - kv 14: tokenizer.ggml.model str = gpt2
llama_model_loader: - kv 15: tokenizer.ggml.pre str = qwen2
llama_model_loader: - kv 16: tokenizer.ggml.tokens arr[str,152064] = ["!", """, "#", "$", "%", "&", "'", ...
llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv 18: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 151646
llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151643
llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643
llama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = true
llama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false
llama_model_loader: - kv 24: tokenizer.chat_template str = {% if not add_generation_prompt is de...
llama_model_loader: - kv 25: general.quantization_version u32 = 2
llama_model_loader: - type f32: 241 tensors
llama_model_loader: - type q4_K: 289 tensors
llama_model_loader: - type q6_K: 49 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type = Q4_K - Medium
print_info: file size = 8.37 GiB (4.87 BPW)
load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
load: printing all EOG tokens:
load: - 151643 ('<｜end▁of▁sentence｜>')
load: - 151662 ('<|fim_pad|>')
load: - 151663 ('<|repo_name|>')
load: - 151664 ('<|file_sep|>')
load: special tokens cache size = 22
load: token to piece cache size = 0.9310 MB
print_info: arch = qwen2
print_info: vocab_only = 0
print_info: no_alloc = 0
print_info: n_ctx_train = 131072
print_info: n_embd = 5120
print_info: n_embd_inp = 5120
print_info: n_layer = 48
print_info: n_head = 40
print_info: n_head_kv = 8
print_info: n_rot = 128
print_info: n_swa = 0
print_info: is_swa_any = 0
print_info: n_embd_head_k = 128
print_info: n_embd_head_v = 128
print_info: n_gqa = 5
print_info: n_embd_k_gqa = 1024
print_info: n_embd_v_gqa = 1024
print_info: f_norm_eps = 0.0e+00
print_info: f_norm_rms_eps = 1.0e-05
print_info: f_clamp_kqv = 0.0e+00
print_info: f_max_alibi_bias = 0.0e+00
print_info: f_logit_scale = 0.0e+00
print_info: f_attn_scale = 0.0e+00
print_info: n_ff = 13824
print_info: n_expert = 0
print_info: n_expert_used = 0
print_info: n_expert_groups = 0
print_info: n_group_used = 0
print_info: causal attn = 1
print_info: pooling type = -1
print_info: rope type = 2
print_info: rope scaling = linear
print_info: freq_base_train = 1000000.0
print_info: freq_scale_train = 1
print_info: n_ctx_orig_yarn = 131072
print_info: rope_yarn_log_mul= 0.0000
print_info: rope_finetuned = unknown
print_info: model type = 14B
print_info: model params = 14.77 B
print_info: general.name = DeepSeek R1 Distill Qwen 14B
print_info: vocab type = BPE
print_info: n_vocab = 152064
print_info: n_merges = 151387
print_info: BOS token = 151646 '<｜begin▁of▁sentence｜>'
print_info: EOS token = 151643 '<｜end▁of▁sentence｜>'
print_info: EOT token = 151643 '<｜end▁of▁sentence｜>'
print_info: PAD token = 151643 '<｜end▁of▁sentence｜>'
print_info: LF token = 198 'Ċ'
print_info: FIM PRE token = 151659 '<|fim_prefix|>'
print_info: FIM SUF token = 151661 '<|fim_suffix|>'
print_info: FIM MID token = 151660 '<|fim_middle|>'
print_info: FIM PAD token = 151662 '<|fim_pad|>'
print_info: FIM REP token = 151663 '<|repo_name|>'
print_info: FIM SEP token = 151664 '<|file_sep|>'
print_info: EOG token = 151643 '<｜end▁of▁sentence｜>'
print_info: EOG token = 151662 '<|fim_pad|>'
print_info: EOG token = 151663 '<|repo_name|>'
print_info: EOG token = 151664 '<|file_sep|>'
print_info: max token length = 256
load_tensors: loading model tensors, this can take a while... (mmap = false)
load_tensors: CPU model buffer size = 8566.04 MiB
llama_context: constructing llama_context
llama_context: n_seq_max = 1
llama_context: n_ctx = 4096
llama_context: n_ctx_seq = 4096
llama_context: n_batch = 512
llama_context: n_ubatch = 512
llama_context: causal_attn = 1
llama_context: flash_attn = auto
llama_context: kv_unified = false
llama_context: freq_base = 1000000.0
llama_context: freq_scale = 1
llama_context: n_ctx_seq (4096) < n_ctx_train (131072) -- the full capacity of the model will not be utilized
llama_context: CPU output buffer size = 0.60 MiB
llama_kv_cache: CPU KV buffer size = 768.00 MiB
llama_kv_cache: size = 768.00 MiB ( 4096 cells, 48 layers, 1/1 seqs), K (f16): 384.00 MiB, V (f16): 384.00 MiB
llama_context: Flash Attention was auto, set to enabled
llama_context: CPU compute buffer size = 349.01 MiB
llama_context: graph nodes = 1639
llama_context: graph splits = 1
time=2026-02-17T16:22:05.502+01:00 level=INFO source=server.go:1388 msg="llama runner started in 22.22 seconds"
time=2026-02-17T16:22:05.502+01:00 level=INFO source=sched.go:540 msg="loaded runners" count=1
time=2026-02-17T16:22:05.502+01:00 level=INFO source=server.go:1350 msg="waiting for llama runner to start responding"
time=2026-02-17T16:22:05.503+01:00 level=INFO source=server.go:1388 msg="llama runner started in 22.22 seconds"

@TiiStyle commented on GitHub (Feb 17, 2026): Here are the combined logs for: num_gpu: 4, 2, 1 and 0. time=2026-02-17T16:09:47.967+01:00 level=INFO source=routes.go:1663 msg="server config" env="map[CUDA_VISIBLE_DEVICES:0 GGML_VK_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:4096 OLLAMA_DEBUG:INFO OLLAMA_EDITOR: OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:F:\\OllamaModels OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NO_CLOUD:false OLLAMA_NUM_PARALLEL:1 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_REMOTES:[ollama.com] OLLAMA_SCHED_SPREAD:false OLLAMA_VULKAN:false ROCR_VISIBLE_DEVICES:]" time=2026-02-17T16:09:48.077+01:00 level=INFO source=routes.go:1665 msg="Ollama cloud disabled: false" time=2026-02-17T16:09:48.095+01:00 level=INFO source=images.go:473 msg="total blobs: 41" time=2026-02-17T16:09:48.102+01:00 level=INFO source=images.go:480 msg="total unused blobs removed: 0" time=2026-02-17T16:09:48.109+01:00 level=INFO source=routes.go:1718 msg="Listening on 127.0.0.1:11434 (version 0.16.2)" time=2026-02-17T16:09:48.124+01:00 level=INFO source=runner.go:67 msg="discovering available GPUs..." time=2026-02-17T16:09:48.197+01:00 level=WARN source=runner.go:485 msg="user overrode visible devices" CUDA_VISIBLE_DEVICES=0 time=2026-02-17T16:09:48.197+01:00 level=WARN source=runner.go:489 msg="if GPUs are not correctly discovered, unset and try again" time=2026-02-17T16:09:48.211+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 33827" time=2026-02-17T16:09:48.573+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 33848" time=2026-02-17T16:09:50.055+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 33866" time=2026-02-17T16:09:50.457+01:00 level=INFO source=runner.go:106 msg="experimental Vulkan support disabled. To enable, set OLLAMA_VULKAN=1" time=2026-02-17T16:09:50.460+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 33877" time=2026-02-17T16:09:50.460+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 33878" time=2026-02-17T16:09:50.792+01:00 level=INFO source=types.go:42 msg="inference compute" id=GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 filter_id="" library=CUDA compute=5.2 name=CUDA0 description="NVIDIA GeForce GTX 980" libdirs=ollama,cuda_v12 driver=13.0 pci_id=0000:01:00.0 type=discrete total="4.0 GiB" available="3.7 GiB" time=2026-02-17T16:09:50.792+01:00 level=INFO source=routes.go:1768 msg="vram-based default context" total_vram="4.0 GiB" default_num_ctx=4096 [GIN] 2026/02/17 - 16:09:50 | 200 | 993.2µs | 127.0.0.1 | GET "/api/version" [GIN] 2026/02/17 - 16:09:50 | 200 | 993.2µs | 127.0.0.1 | GET "/api/version" [GIN] 2026/02/17 - 16:09:50 | 200 | 0s | 127.0.0.1 | GET "/api/version" [GIN] 2026/02/17 - 16:09:50 | 200 | 7.9405ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:09:50 | 200 | 140.1508ms | 127.0.0.1 | POST "/api/show" [GIN] 2026/02/17 - 16:09:50 | 401 | 192.2531ms | 127.0.0.1 | POST "/api/me" [GIN] 2026/02/17 - 16:09:51 | 401 | 207.5809ms | 127.0.0.1 | POST "/api/me" [GIN] 2026/02/17 - 16:09:57 | 200 | 112.0475ms | 127.0.0.1 | POST "/api/show" [GIN] 2026/02/17 - 16:10:04 | 200 | 6.1967ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:10:04 | 200 | 123.9063ms | 127.0.0.1 | POST "/api/show" [GIN] 2026/02/17 - 16:10:04 | 200 | 116.1573ms | 127.0.0.1 | POST "/api/show" time=2026-02-17T16:10:04.333+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 33916" time=2026-02-17T16:10:04.615+01:00 level=INFO source=cpu_windows.go:148 msg=packages count=1 time=2026-02-17T16:10:04.615+01:00 level=INFO source=cpu_windows.go:195 msg="" package=0 cores=4 efficiency=0 threads=8 llama_model_loader: loaded meta data with 26 key-value pairs and 579 tensors from F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e (version GGUF V3 (latest)) llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output. llama_model_loader: - kv 0: general.architecture str = qwen2 llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.name str = DeepSeek R1 Distill Qwen 14B llama_model_loader: - kv 3: general.basename str = DeepSeek-R1-Distill-Qwen llama_model_loader: - kv 4: general.size_label str = 14B llama_model_loader: - kv 5: qwen2.block_count u32 = 48 llama_model_loader: - kv 6: qwen2.context_length u32 = 131072 llama_model_loader: - kv 7: qwen2.embedding_length u32 = 5120 llama_model_loader: - kv 8: qwen2.feed_forward_length u32 = 13824 llama_model_loader: - kv 9: qwen2.attention.head_count u32 = 40 llama_model_loader: - kv 10: qwen2.attention.head_count_kv u32 = 8 llama_model_loader: - kv 11: qwen2.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 12: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 13: general.file_type u32 = 15 llama_model_loader: - kv 14: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 15: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 16: tokenizer.ggml.tokens arr[str,152064] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 18: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 151646 llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151643 llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 24: tokenizer.chat_template str = {% if not add_generation_prompt is de... llama_model_loader: - kv 25: general.quantization_version u32 = 2 llama_model_loader: - type f32: 241 tensors llama_model_loader: - type q4_K: 289 tensors llama_model_loader: - type q6_K: 49 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 8.37 GiB (4.87 BPW) load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect load: printing all EOG tokens: load: - 151643 ('<｜end▁of▁sentence｜>') load: - 151662 ('<|fim_pad|>') load: - 151663 ('<|repo_name|>') load: - 151664 ('<|file_sep|>') load: special tokens cache size = 22 load: token to piece cache size = 0.9310 MB print_info: arch = qwen2 print_info: vocab_only = 1 print_info: no_alloc = 0 print_info: model type = ?B print_info: model params = 14.77 B print_info: general.name = DeepSeek R1 Distill Qwen 14B print_info: vocab type = BPE print_info: n_vocab = 152064 print_info: n_merges = 151387 print_info: BOS token = 151646 '<｜begin▁of▁sentence｜>' print_info: EOS token = 151643 '<｜end▁of▁sentence｜>' print_info: EOT token = 151643 '<｜end▁of▁sentence｜>' print_info: PAD token = 151643 '<｜end▁of▁sentence｜>' print_info: LF token = 198 'Ċ' print_info: FIM PRE token = 151659 '<|fim_prefix|>' print_info: FIM SUF token = 151661 '<|fim_suffix|>' print_info: FIM MID token = 151660 '<|fim_middle|>' print_info: FIM PAD token = 151662 '<|fim_pad|>' print_info: FIM REP token = 151663 '<|repo_name|>' print_info: FIM SEP token = 151664 '<|file_sep|>' print_info: EOG token = 151643 '<｜end▁of▁sentence｜>' print_info: EOG token = 151662 '<|fim_pad|>' print_info: EOG token = 151663 '<|repo_name|>' print_info: EOG token = 151664 '<|file_sep|>' print_info: max token length = 256 llama_model_load: vocab only - skipping tensors time=2026-02-17T16:10:04.945+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --model F:\\OllamaModels\\blobs\\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e --port 33927" time=2026-02-17T16:10:04.946+01:00 level=INFO source=sched.go:466 msg="system memory" total="15.9 GiB" free="11.2 GiB" free_swap="19.6 GiB" time=2026-02-17T16:10:04.946+01:00 level=INFO source=sched.go:473 msg="gpu memory" id=GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 library=CUDA available="3.3 GiB" free="3.7 GiB" minimum="457.0 MiB" overhead="0 B" time=2026-02-17T16:10:04.946+01:00 level=INFO source=server.go:498 msg="loading model" "model layers"=49 requested=4 time=2026-02-17T16:10:04.947+01:00 level=INFO source=device.go:240 msg="model weights" device=CUDA0 size="665.7 MiB" time=2026-02-17T16:10:04.947+01:00 level=INFO source=device.go:245 msg="model weights" device=CPU size="7.3 GiB" time=2026-02-17T16:10:04.947+01:00 level=INFO source=device.go:251 msg="kv cache" device=CUDA0 size="64.0 MiB" time=2026-02-17T16:10:04.947+01:00 level=INFO source=device.go:256 msg="kv cache" device=CPU size="704.0 MiB" time=2026-02-17T16:10:04.947+01:00 level=INFO source=device.go:262 msg="compute graph" device=CUDA0 size="916.1 MiB" time=2026-02-17T16:10:04.947+01:00 level=INFO source=device.go:272 msg="total memory" size="9.6 GiB" time=2026-02-17T16:10:04.989+01:00 level=INFO source=runner.go:965 msg="starting go runner" load_backend: loaded CPU backend from C:\Users\robob\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no ggml_cuda_init: found 1 CUDA devices: Device 0: NVIDIA GeForce GTX 980, compute capability 5.2, VMM: yes, ID: GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 load_backend: loaded CUDA backend from C:\Users\robob\AppData\Local\Programs\Ollama\lib\ollama\cuda_v12\ggml-cuda.dll time=2026-02-17T16:10:05.177+01:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 CUDA.0.ARCHS=500,520,600,610,700,750,800,860,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(clang) time=2026-02-17T16:10:05.178+01:00 level=INFO source=runner.go:1001 msg="Server listening on 127.0.0.1:33927" time=2026-02-17T16:10:05.185+01:00 level=INFO source=runner.go:895 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Auto KvSize:4096 KvCacheType: NumThreads:4 GPULayers:4[ID:GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 Layers:4(44..47)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2026-02-17T16:10:05.185+01:00 level=INFO source=server.go:1350 msg="waiting for llama runner to start responding" time=2026-02-17T16:10:05.185+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server loading model" ggml_backend_cuda_device_get_memory device GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 utilizing NVML memory reporting free: 3986210816 total: 4294967296 llama_model_load_from_file_impl: using device CUDA0 (NVIDIA GeForce GTX 980) (0000:01:00.0) - 3801 MiB free llama_model_loader: loaded meta data with 26 key-value pairs and 579 tensors from F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e (version GGUF V3 (latest)) llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output. llama_model_loader: - kv 0: general.architecture str = qwen2 llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.name str = DeepSeek R1 Distill Qwen 14B llama_model_loader: - kv 3: general.basename str = DeepSeek-R1-Distill-Qwen llama_model_loader: - kv 4: general.size_label str = 14B llama_model_loader: - kv 5: qwen2.block_count u32 = 48 llama_model_loader: - kv 6: qwen2.context_length u32 = 131072 llama_model_loader: - kv 7: qwen2.embedding_length u32 = 5120 llama_model_loader: - kv 8: qwen2.feed_forward_length u32 = 13824 llama_model_loader: - kv 9: qwen2.attention.head_count u32 = 40 llama_model_loader: - kv 10: qwen2.attention.head_count_kv u32 = 8 llama_model_loader: - kv 11: qwen2.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 12: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 13: general.file_type u32 = 15 llama_model_loader: - kv 14: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 15: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 16: tokenizer.ggml.tokens arr[str,152064] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 18: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 151646 llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151643 llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 24: tokenizer.chat_template str = {% if not add_generation_prompt is de... llama_model_loader: - kv 25: general.quantization_version u32 = 2 llama_model_loader: - type f32: 241 tensors llama_model_loader: - type q4_K: 289 tensors llama_model_loader: - type q6_K: 49 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 8.37 GiB (4.87 BPW) load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect load: printing all EOG tokens: load: - 151643 ('<｜end▁of▁sentence｜>') load: - 151662 ('<|fim_pad|>') load: - 151663 ('<|repo_name|>') load: - 151664 ('<|file_sep|>') load: special tokens cache size = 22 load: token to piece cache size = 0.9310 MB print_info: arch = qwen2 print_info: vocab_only = 0 print_info: no_alloc = 0 print_info: n_ctx_train = 131072 print_info: n_embd = 5120 print_info: n_embd_inp = 5120 print_info: n_layer = 48 print_info: n_head = 40 print_info: n_head_kv = 8 print_info: n_rot = 128 print_info: n_swa = 0 print_info: is_swa_any = 0 print_info: n_embd_head_k = 128 print_info: n_embd_head_v = 128 print_info: n_gqa = 5 print_info: n_embd_k_gqa = 1024 print_info: n_embd_v_gqa = 1024 print_info: f_norm_eps = 0.0e+00 print_info: f_norm_rms_eps = 1.0e-05 print_info: f_clamp_kqv = 0.0e+00 print_info: f_max_alibi_bias = 0.0e+00 print_info: f_logit_scale = 0.0e+00 print_info: f_attn_scale = 0.0e+00 print_info: n_ff = 13824 print_info: n_expert = 0 print_info: n_expert_used = 0 print_info: n_expert_groups = 0 print_info: n_group_used = 0 print_info: causal attn = 1 print_info: pooling type = -1 print_info: rope type = 2 print_info: rope scaling = linear print_info: freq_base_train = 1000000.0 print_info: freq_scale_train = 1 print_info: n_ctx_orig_yarn = 131072 print_info: rope_yarn_log_mul= 0.0000 print_info: rope_finetuned = unknown print_info: model type = 14B print_info: model params = 14.77 B print_info: general.name = DeepSeek R1 Distill Qwen 14B print_info: vocab type = BPE print_info: n_vocab = 152064 print_info: n_merges = 151387 print_info: BOS token = 151646 '<｜begin▁of▁sentence｜>' print_info: EOS token = 151643 '<｜end▁of▁sentence｜>' print_info: EOT token = 151643 '<｜end▁of▁sentence｜>' print_info: PAD token = 151643 '<｜end▁of▁sentence｜>' print_info: LF token = 198 'Ċ' print_info: FIM PRE token = 151659 '<|fim_prefix|>' print_info: FIM SUF token = 151661 '<|fim_suffix|>' print_info: FIM MID token = 151660 '<|fim_middle|>' print_info: FIM PAD token = 151662 '<|fim_pad|>' print_info: FIM REP token = 151663 '<|repo_name|>' print_info: FIM SEP token = 151664 '<|file_sep|>' print_info: EOG token = 151643 '<｜end▁of▁sentence｜>' print_info: EOG token = 151662 '<|fim_pad|>' print_info: EOG token = 151663 '<|repo_name|>' print_info: EOG token = 151664 '<|file_sep|>' print_info: max token length = 256 load_tensors: loading model tensors, this can take a while... (mmap = false) ggml_cuda_host_malloc: failed to allocate 7482.73 MiB of pinned memory: out of memory load_tensors: offloading 4 repeating layers to GPU load_tensors: offloaded 4/49 layers to GPU load_tensors: CPU model buffer size = 417.66 MiB load_tensors: CUDA0 model buffer size = 665.66 MiB load_tensors: CPU model buffer size = 7482.73 MiB llama_context: constructing llama_context llama_context: n_seq_max = 1 llama_context: n_ctx = 4096 llama_context: n_ctx_seq = 4096 llama_context: n_batch = 512 llama_context: n_ubatch = 512 llama_context: causal_attn = 1 llama_context: flash_attn = auto llama_context: kv_unified = false llama_context: freq_base = 1000000.0 llama_context: freq_scale = 1 llama_context: n_ctx_seq (4096) < n_ctx_train (131072) -- the full capacity of the model will not be utilized llama_context: CPU output buffer size = 0.60 MiB llama_kv_cache: CPU KV buffer size = 704.00 MiB llama_kv_cache: CUDA0 KV buffer size = 64.00 MiB llama_kv_cache: size = 768.00 MiB ( 4096 cells, 48 layers, 1/1 seqs), K (f16): 384.00 MiB, V (f16): 384.00 MiB llama_context: Flash Attention was auto, set to enabled graph_reserve: failed to allocate compute buffers Exception 0xc0000005 0x0 0x1f374b48f70 0x7ffceb1fe53e PC=0x7ffceb1fe53e signal arrived during external code execution runtime.cgocall(0x7ff793e6e2b0, 0xc00014dc00) runtime/cgocall.go:167 +0x3e fp=0xc00014dbd8 sp=0xc00014db70 pc=0x7ff792fd243e github.com/ollama/ollama/llama._Cfunc_llama_init_from_model(0x1f96f41fca0, {0x1000, 0x200, 0x200, 0x1, 0x4, 0x4, 0xffffffff, 0xffffffff, 0xffffffff, ...}) _cgo_gotypes.go:759 +0x54 fp=0xc00014dc00 sp=0xc00014dbd8 pc=0x7ff793479774 github.com/ollama/ollama/llama.NewContextWithModel.func1(...) github.com/ollama/ollama/llama/llama.go:322 github.com/ollama/ollama/llama.NewContextWithModel(0xc000346d38, {{0x1000, 0x200, 0x200, 0x1, 0x4, 0x4, 0xffffffff, 0xffffffff, 0xffffffff, ...}}) github.com/ollama/ollama/llama/llama.go:322 +0x158 fp=0xc00014dda0 sp=0xc00014dc00 pc=0x7ff79347dd58 github.com/ollama/ollama/runner/llamarunner.(*Server).loadModel(0xc0003d9b80, {{0xc0000f58f8, 0x1, 0x1}, 0x4, 0x0, 0x0, {0xc0000f58f0, 0x1, 0x2}, ...}, ...) github.com/ollama/ollama/runner/llamarunner/runner.go:847 +0x178 fp=0xc00014dee8 sp=0xc00014dda0 pc=0x7ff79352c938 github.com/ollama/ollama/runner/llamarunner.(*Server).load.gowrap2() github.com/ollama/ollama/runner/llamarunner/runner.go:934 +0x114 fp=0xc00014dfe0 sp=0xc00014dee8 pc=0x7ff79352db54 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc00014dfe8 sp=0xc00014dfe0 pc=0x7ff792fdd9a1 created by github.com/ollama/ollama/runner/llamarunner.(*Server).load in goroutine 10 github.com/ollama/ollama/runner/llamarunner/runner.go:934 +0x889 goroutine 1 gp=0xc0000021c0 m=nil [IO wait]: runtime.gopark(0x7ff792fdf1a0?, 0x7ff795180d00?, 0xa0?, 0xf1?, 0xc00046f24c?) runtime/proc.go:435 +0xce fp=0xc00011f630 sp=0xc00011f610 pc=0x7ff792fd598e runtime.netpollblock(0x388?, 0x92f70406?, 0xf7?) runtime/netpoll.go:575 +0xf7 fp=0xc00011f668 sp=0xc00011f630 pc=0x7ff792f9bdf7 internal/poll.runtime_pollWait(0x1f96c69a030, 0x72) runtime/netpoll.go:351 +0x85 fp=0xc00011f688 sp=0xc00011f668 pc=0x7ff792fd4b25 internal/poll.(*pollDesc).wait(0x7ff79306a8f3?, 0x0?, 0x0) internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc00011f6b0 sp=0xc00011f688 pc=0x7ff79306bee7 internal/poll.execIO(0xc00046f1a0, 0xc00011f758) internal/poll/fd_windows.go:177 +0x105 fp=0xc00011f728 sp=0xc00011f6b0 pc=0x7ff79306d345 internal/poll.(*FD).acceptOne(0xc00046f188, 0x394, {0xc0001440f0?, 0xc00011f7b8?, 0x7ff793075005?}, 0xc00011f7ec?) internal/poll/fd_windows.go:946 +0x65 fp=0xc00011f788 sp=0xc00011f728 pc=0x7ff7930718c5 internal/poll.(*FD).Accept(0xc00046f188, 0xc00011f938) internal/poll/fd_windows.go:980 +0x1b6 fp=0xc00011f840 sp=0xc00011f788 pc=0x7ff793071bf6 net.(*netFD).accept(0xc00046f188) net/fd_windows.go:182 +0x4b fp=0xc00011f958 sp=0xc00011f840 pc=0x7ff7930e344b net.(*TCPListener).accept(0xc00069ad40) net/tcpsock_posix.go:159 +0x1b fp=0xc00011f9a8 sp=0xc00011f958 pc=0x7ff7930f99fb net.(*TCPListener).Accept(0xc00069ad40) net/tcpsock.go:380 +0x30 fp=0xc00011f9d8 sp=0xc00011f9a8 pc=0x7ff7930f87b0 net/http.(*onceCloseListener).Accept(0xc00012c480?) <autogenerated>:1 +0x24 fp=0xc00011f9f0 sp=0xc00011f9d8 pc=0x7ff793311ec4 net/http.(*Server).Serve(0xc0001bfd00, {0x7ff7946e21d0, 0xc00069ad40}) net/http/server.go:3424 +0x30c fp=0xc00011fb20 sp=0xc00011f9f0 pc=0x7ff7932e978c github.com/ollama/ollama/runner/llamarunner.Execute({0xc0000a4020, 0x4, 0x6}) github.com/ollama/ollama/runner/llamarunner/runner.go:1002 +0x8f5 fp=0xc00011fcf0 sp=0xc00011fb20 pc=0x7ff79352e515 github.com/ollama/ollama/runner.Execute({0xc0000a4010?, 0x0?, 0x0?}) github.com/ollama/ollama/runner/runner.go:25 +0x1a5 fp=0xc00011fd30 sp=0xc00011fcf0 pc=0x7ff793603ae5 github.com/ollama/ollama/cmd.NewCLI.func3(0xc0001bf800?, {0x7ff7944c4587?, 0x4?, 0x7ff7944c458b?}) github.com/ollama/ollama/cmd/cmd.go:2262 +0x45 fp=0xc00011fd58 sp=0xc00011fd30 pc=0x7ff793dfce65 github.com/spf13/cobra.(*Command).execute(0xc0001c7508, {0xc0003d3e00, 0x4, 0x4}) github.com/spf13/cobra@v1.7.0/command.go:940 +0x85c fp=0xc00011fe78 sp=0xc00011fd58 pc=0x7ff79315e61c github.com/spf13/cobra.(*Command).ExecuteC(0xc0006d4908) github.com/spf13/cobra@v1.7.0/command.go:1068 +0x3a5 fp=0xc00011ff30 sp=0xc00011fe78 pc=0x7ff79315ee65 github.com/spf13/cobra.(*Command).Execute(...) github.com/spf13/cobra@v1.7.0/command.go:992 github.com/spf13/cobra.(*Command).ExecuteContext(...) github.com/spf13/cobra@v1.7.0/command.go:985 main.main() github.com/ollama/ollama/main.go:12 +0x4d fp=0xc00011ff50 sp=0xc00011ff30 pc=0x7ff793dff28d runtime.main() runtime/proc.go:283 +0x27d fp=0xc00011ffe0 sp=0xc00011ff50 pc=0x7ff792fa4ddd runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc00011ffe8 sp=0xc00011ffe0 pc=0x7ff792fdd9a1 goroutine 2 gp=0xc0000028c0 m=nil [force gc (idle)]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000071fa8 sp=0xc000071f88 pc=0x7ff792fd598e runtime.goparkunlock(...) runtime/proc.go:441 runtime.forcegchelper() runtime/proc.go:348 +0xb8 fp=0xc000071fe0 sp=0xc000071fa8 pc=0x7ff792fa50f8 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000071fe8 sp=0xc000071fe0 pc=0x7ff792fdd9a1 created by runtime.init.7 in goroutine 1 runtime/proc.go:336 +0x1a goroutine 3 gp=0xc000002c40 m=nil [GC sweep wait]: runtime.gopark(0x1?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000073f80 sp=0xc000073f60 pc=0x7ff792fd598e runtime.goparkunlock(...) runtime/proc.go:441 runtime.bgsweep(0xc000080000) runtime/mgcsweep.go:316 +0xdf fp=0xc000073fc8 sp=0xc000073f80 pc=0x7ff792f8debf runtime.gcenable.gowrap1() runtime/mgc.go:204 +0x25 fp=0xc000073fe0 sp=0xc000073fc8 pc=0x7ff792f82285 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000073fe8 sp=0xc000073fe0 pc=0x7ff792fdd9a1 created by runtime.gcenable in goroutine 1 runtime/mgc.go:204 +0x66 goroutine 4 gp=0xc000002e00 m=nil [GC scavenge wait]: runtime.gopark(0x10000?, 0x7ff7946ccc38?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000087f78 sp=0xc000087f58 pc=0x7ff792fd598e runtime.goparkunlock(...) runtime/proc.go:441 runtime.(*scavengerState).park(0x7ff7951aa900) runtime/mgcscavenge.go:425 +0x49 fp=0xc000087fa8 sp=0xc000087f78 pc=0x7ff792f8b909 runtime.bgscavenge(0xc000080000) runtime/mgcscavenge.go:658 +0x59 fp=0xc000087fc8 sp=0xc000087fa8 pc=0x7ff792f8be99 runtime.gcenable.gowrap2() runtime/mgc.go:205 +0x25 fp=0xc000087fe0 sp=0xc000087fc8 pc=0x7ff792f82225 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000087fe8 sp=0xc000087fe0 pc=0x7ff792fdd9a1 created by runtime.gcenable in goroutine 1 runtime/mgc.go:205 +0xa5 goroutine 5 gp=0xc000003340 m=nil [finalizer wait]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000089e30 sp=0xc000089e10 pc=0x7ff792fd598e runtime.runfinq() runtime/mfinal.go:196 +0x107 fp=0xc000089fe0 sp=0xc000089e30 pc=0x7ff792f81207 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000089fe8 sp=0xc000089fe0 pc=0x7ff792fdd9a1 created by runtime.createfing in goroutine 1 runtime/mfinal.go:166 +0x3d goroutine 6 gp=0xc000003dc0 m=nil [chan receive]: runtime.gopark(0xc0001f14a0?, 0xc000508018?, 0x60?, 0x5f?, 0x7ff7930cc148?) runtime/proc.go:435 +0xce fp=0xc000075f18 sp=0xc000075ef8 pc=0x7ff792fd598e runtime.chanrecv(0xc00008c380, 0x0, 0x1) runtime/chan.go:664 +0x445 fp=0xc000075f90 sp=0xc000075f18 pc=0x7ff792f72d45 runtime.chanrecv1(0x7ff792fa4f40?, 0xc000075f76?) runtime/chan.go:506 +0x12 fp=0xc000075fb8 sp=0xc000075f90 pc=0x7ff792f728d2 runtime.unique_runtime_registerUniqueMapCleanup.func2(...) runtime/mgc.go:1796 runtime.unique_runtime_registerUniqueMapCleanup.gowrap1() runtime/mgc.go:1799 +0x2f fp=0xc000075fe0 sp=0xc000075fb8 pc=0x7ff792f854af runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000075fe8 sp=0xc000075fe0 pc=0x7ff792fdd9a1 created by unique.runtime_registerUniqueMapCleanup in goroutine 1 runtime/mgc.go:1794 +0x85 goroutine 7 gp=0xc0003da540 m=nil [GC worker (idle)]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000083f38 sp=0xc000083f18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008d960) runtime/mgc.go:1423 +0xe9 fp=0xc000083fc8 sp=0xc000083f38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc000083fe0 sp=0xc000083fc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000083fe8 sp=0xc000083fe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 18 gp=0xc0001061c0 m=nil [GC worker (idle)]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000113f38 sp=0xc000113f18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008d960) runtime/mgc.go:1423 +0xe9 fp=0xc000113fc8 sp=0xc000113f38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc000113fe0 sp=0xc000113fc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000113fe8 sp=0xc000113fe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 34 gp=0xc000484000 m=nil [GC worker (idle)]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc00010ff38 sp=0xc00010ff18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008d960) runtime/mgc.go:1423 +0xe9 fp=0xc00010ffc8 sp=0xc00010ff38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc00010ffe0 sp=0xc00010ffc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc00010ffe8 sp=0xc00010ffe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 35 gp=0xc0004841c0 m=nil [GC worker (idle)]: runtime.gopark(0x4b423cff1788?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000111f38 sp=0xc000111f18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008d960) runtime/mgc.go:1423 +0xe9 fp=0xc000111fc8 sp=0xc000111f38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc000111fe0 sp=0xc000111fc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000111fe8 sp=0xc000111fe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 36 gp=0xc000484380 m=nil [GC worker (idle)]: runtime.gopark(0x4b423cf7000c?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc00048bf38 sp=0xc00048bf18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008d960) runtime/mgc.go:1423 +0xe9 fp=0xc00048bfc8 sp=0xc00048bf38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc00048bfe0 sp=0xc00048bfc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc00048bfe8 sp=0xc00048bfe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 8 gp=0xc0003da700 m=nil [GC worker (idle)]: runtime.gopark(0x4b423cff1788?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000085f38 sp=0xc000085f18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008d960) runtime/mgc.go:1423 +0xe9 fp=0xc000085fc8 sp=0xc000085f38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc000085fe0 sp=0xc000085fc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000085fe8 sp=0xc000085fe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 19 gp=0xc000106380 m=nil [GC worker (idle)]: runtime.gopark(0x4b423cff1788?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000115f38 sp=0xc000115f18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008d960) runtime/mgc.go:1423 +0xe9 fp=0xc000115fc8 sp=0xc000115f38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc000115fe0 sp=0xc000115fc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000115fe8 sp=0xc000115fe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 37 gp=0xc000484540 m=nil [GC worker (idle)]: runtime.gopark(0x7ff7951fc600?, 0x1?, 0x48?, 0x5c?, 0x0?) runtime/proc.go:435 +0xce fp=0xc00048df38 sp=0xc00048df18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008d960) runtime/mgc.go:1423 +0xe9 fp=0xc00048dfc8 sp=0xc00048df38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc00048dfe0 sp=0xc00048dfc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc00048dfe8 sp=0xc00048dfe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 9 gp=0xc000484c40 m=nil [sync.WaitGroup.Wait]: runtime.gopark(0x0?, 0x0?, 0x60?, 0x3e?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000487e20 sp=0xc000487e00 pc=0x7ff792fd598e runtime.goparkunlock(...) runtime/proc.go:441 runtime.semacquire1(0xc0003d9ba0, 0x0, 0x1, 0x0, 0x18) runtime/sema.go:188 +0x22f fp=0xc000487e88 sp=0xc000487e20 pc=0x7ff792fb750f sync.runtime_SemacquireWaitGroup(0x0?) runtime/sema.go:110 +0x25 fp=0xc000487ec0 sp=0xc000487e88 pc=0x7ff792fd6f85 sync.(*WaitGroup).Wait(0x0?) sync/waitgroup.go:118 +0x48 fp=0xc000487ee8 sp=0xc000487ec0 pc=0x7ff792feb988 github.com/ollama/ollama/runner/llamarunner.(*Server).run(0xc0003d9b80, {0x7ff7946e4a30, 0xc00033f770}) github.com/ollama/ollama/runner/llamarunner/runner.go:360 +0x4b fp=0xc000487fb8 sp=0xc000487ee8 pc=0x7ff79352922b github.com/ollama/ollama/runner/llamarunner.Execute.gowrap1() github.com/ollama/ollama/runner/llamarunner/runner.go:981 +0x28 fp=0xc000487fe0 sp=0xc000487fb8 pc=0x7ff79352e788 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000487fe8 sp=0xc000487fe0 pc=0x7ff792fdd9a1 created by github.com/ollama/ollama/runner/llamarunner.Execute in goroutine 1 github.com/ollama/ollama/runner/llamarunner/runner.go:981 +0x4c5 goroutine 10 gp=0xc000484e00 m=nil [IO wait]: runtime.gopark(0x0?, 0xc00046f420?, 0xc8?, 0xf4?, 0xc00046f4cc?) runtime/proc.go:435 +0xce fp=0xc00004b8c8 sp=0xc00004b8a8 pc=0x7ff792fd598e runtime.netpollblock(0x390?, 0x92f70406?, 0xf7?) runtime/netpoll.go:575 +0xf7 fp=0xc00004b900 sp=0xc00004b8c8 pc=0x7ff792f9bdf7 internal/poll.runtime_pollWait(0x1f96c699f18, 0x72) runtime/netpoll.go:351 +0x85 fp=0xc00004b920 sp=0xc00004b900 pc=0x7ff792fd4b25 internal/poll.(*pollDesc).wait(0x7ff79319d5f7?, 0xc00004b970?, 0x0) internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc00004b948 sp=0xc00004b920 pc=0x7ff79306bee7 internal/poll.execIO(0xc00046f420, 0x7ff794546098) internal/poll/fd_windows.go:177 +0x105 fp=0xc00004b9c0 sp=0xc00004b948 pc=0x7ff79306d345 internal/poll.(*FD).Read(0xc00046f408, {0xc000091000, 0x1000, 0x1000}) internal/poll/fd_windows.go:438 +0x29b fp=0xc00004ba60 sp=0xc00004b9c0 pc=0x7ff79306e01b net.(*netFD).Read(0xc00046f408, {0xc000091000?, 0xc00004bad0?, 0x7ff79306c3a5?}) net/fd_posix.go:55 +0x25 fp=0xc00004baa8 sp=0xc00004ba60 pc=0x7ff7930e1325 net.(*conn).Read(0xc000078740, {0xc000091000?, 0x0?, 0x0?}) net/net.go:194 +0x45 fp=0xc00004baf0 sp=0xc00004baa8 pc=0x7ff7930f0a45 net/http.(*connReader).Read(0xc000246090, {0xc000091000, 0x1000, 0x1000}) net/http/server.go:798 +0x159 fp=0xc00004bb40 sp=0xc00004baf0 pc=0x7ff7932de639 bufio.(*Reader).fill(0xc000213ec0) bufio/bufio.go:113 +0x103 fp=0xc00004bb78 sp=0xc00004bb40 pc=0x7ff793107963 bufio.(*Reader).Peek(0xc000213ec0, 0x4) bufio/bufio.go:152 +0x53 fp=0xc00004bb98 sp=0xc00004bb78 pc=0x7ff793107a93 net/http.(*conn).serve(0xc00012c480, {0x7ff7946e49f8, 0xc0003243f0}) net/http/server.go:2137 +0x785 fp=0xc00004bfb8 sp=0xc00004bb98 pc=0x7ff7932e4425 net/http.(*Server).Serve.gowrap3() net/http/server.go:3454 +0x28 fp=0xc00004bfe0 sp=0xc00004bfb8 pc=0x7ff7932e9b88 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc00004bfe8 sp=0xc00004bfe0 pc=0x7ff792fdd9a1 created by net/http.(*Server).Serve in goroutine 1 net/http/server.go:3454 +0x485 rax 0xfffffff9fdb6ed00 rbx 0x1f976fda270 rcx 0x1f96f61a8a0 rdx 0x0 rdi 0x0 rsi 0x1f96f61cc80 rbp 0xf1506ffbd0 rsp 0xf1506ff920 r8 0x1f947061080 r9 0x1 r10 0x8000 r11 0xf1506ff930 r12 0x667 r13 0x1218 r14 0x1 r15 0x1f96f41fca0 rip 0x7ffceb1fe53e rflags 0x10287 cs 0x33 fs 0x53 gs 0x2b time=2026-02-17T16:10:27.322+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server error" time=2026-02-17T16:10:28.074+01:00 level=INFO source=sched.go:493 msg="Load failed" model=F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e error="llama runner process has terminated: exit status 2" [GIN] 2026/02/17 - 16:10:28 | 500 | 23.8713951s | 127.0.0.1 | POST "/api/chat" [GIN] 2026/02/17 - 16:10:34 | 200 | 12.5133ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:11:04 | 200 | 5.2293ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:11:34 | 200 | 4.2837ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:12:04 | 200 | 12.7989ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:12:34 | 200 | 4.2427ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:13:04 | 200 | 5.9933ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:13:34 | 200 | 5.0279ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:14:04 | 200 | 11.724ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:14:34 | 200 | 4.7326ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:15:04 | 200 | 5.4637ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:15:34 | 200 | 4.426ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:16:04 | 200 | 5.009ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:16:34 | 200 | 4.6721ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:17:04 | 200 | 5.6033ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:17:34 | 200 | 4.5266ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:18:04 | 200 | 4.8485ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:18:34 | 200 | 4.644ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:18:59 | 200 | 0s | 127.0.0.1 | HEAD "/" [GIN] 2026/02/17 - 16:19:05 | 200 | 5.0163ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:19:36 | 200 | 4.1555ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:19:40 | 200 | 0s | 127.0.0.1 | HEAD "/" [GIN] 2026/02/17 - 16:19:40 | 200 | 52.8589ms | 127.0.0.1 | POST "/api/create" [GIN] 2026/02/17 - 16:19:51 | 200 | 0s | 127.0.0.1 | HEAD "/" [GIN] 2026/02/17 - 16:19:51 | 200 | 46.3741ms | 127.0.0.1 | POST "/api/create" [GIN] 2026/02/17 - 16:20:07 | 200 | 5.1281ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:20:08 | 200 | 0s | 127.0.0.1 | HEAD "/" [GIN] 2026/02/17 - 16:20:08 | 200 | 38.7198ms | 127.0.0.1 | POST "/api/create" [GIN] 2026/02/17 - 16:20:11 | 200 | 1.0024ms | 127.0.0.1 | GET "/api/version" [GIN] 2026/02/17 - 16:20:11 | 200 | 16.2654ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:20:11 | 401 | 192.0709ms | 127.0.0.1 | POST "/api/me" [GIN] 2026/02/17 - 16:20:13 | 200 | 123.5651ms | 127.0.0.1 | POST "/api/show" [GIN] 2026/02/17 - 16:20:16 | 200 | 120.6338ms | 127.0.0.1 | POST "/api/show" [GIN] 2026/02/17 - 16:20:16 | 200 | 116.2417ms | 127.0.0.1 | POST "/api/show" time=2026-02-17T16:20:17.006+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 34051" time=2026-02-17T16:20:17.283+01:00 level=INFO source=cpu_windows.go:148 msg=packages count=1 time=2026-02-17T16:20:17.283+01:00 level=INFO source=cpu_windows.go:195 msg="" package=0 cores=4 efficiency=0 threads=8 llama_model_loader: loaded meta data with 26 key-value pairs and 579 tensors from F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e (version GGUF V3 (latest)) llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output. llama_model_loader: - kv 0: general.architecture str = qwen2 llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.name str = DeepSeek R1 Distill Qwen 14B llama_model_loader: - kv 3: general.basename str = DeepSeek-R1-Distill-Qwen llama_model_loader: - kv 4: general.size_label str = 14B llama_model_loader: - kv 5: qwen2.block_count u32 = 48 llama_model_loader: - kv 6: qwen2.context_length u32 = 131072 llama_model_loader: - kv 7: qwen2.embedding_length u32 = 5120 llama_model_loader: - kv 8: qwen2.feed_forward_length u32 = 13824 llama_model_loader: - kv 9: qwen2.attention.head_count u32 = 40 llama_model_loader: - kv 10: qwen2.attention.head_count_kv u32 = 8 llama_model_loader: - kv 11: qwen2.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 12: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 13: general.file_type u32 = 15 llama_model_loader: - kv 14: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 15: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 16: tokenizer.ggml.tokens arr[str,152064] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 18: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 151646 llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151643 llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 24: tokenizer.chat_template str = {% if not add_generation_prompt is de... llama_model_loader: - kv 25: general.quantization_version u32 = 2 llama_model_loader: - type f32: 241 tensors llama_model_loader: - type q4_K: 289 tensors llama_model_loader: - type q6_K: 49 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 8.37 GiB (4.87 BPW) load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect load: printing all EOG tokens: load: - 151643 ('<｜end▁of▁sentence｜>') load: - 151662 ('<|fim_pad|>') load: - 151663 ('<|repo_name|>') load: - 151664 ('<|file_sep|>') load: special tokens cache size = 22 load: token to piece cache size = 0.9310 MB print_info: arch = qwen2 print_info: vocab_only = 1 print_info: no_alloc = 0 print_info: model type = ?B print_info: model params = 14.77 B print_info: general.name = DeepSeek R1 Distill Qwen 14B print_info: vocab type = BPE print_info: n_vocab = 152064 print_info: n_merges = 151387 print_info: BOS token = 151646 '<｜begin▁of▁sentence｜>' print_info: EOS token = 151643 '<｜end▁of▁sentence｜>' print_info: EOT token = 151643 '<｜end▁of▁sentence｜>' print_info: PAD token = 151643 '<｜end▁of▁sentence｜>' print_info: LF token = 198 'Ċ' print_info: FIM PRE token = 151659 '<|fim_prefix|>' print_info: FIM SUF token = 151661 '<|fim_suffix|>' print_info: FIM MID token = 151660 '<|fim_middle|>' print_info: FIM PAD token = 151662 '<|fim_pad|>' print_info: FIM REP token = 151663 '<|repo_name|>' print_info: FIM SEP token = 151664 '<|file_sep|>' print_info: EOG token = 151643 '<｜end▁of▁sentence｜>' print_info: EOG token = 151662 '<|fim_pad|>' print_info: EOG token = 151663 '<|repo_name|>' print_info: EOG token = 151664 '<|file_sep|>' print_info: max token length = 256 llama_model_load: vocab only - skipping tensors time=2026-02-17T16:20:17.602+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --model F:\\OllamaModels\\blobs\\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e --port 34062" time=2026-02-17T16:20:17.603+01:00 level=INFO source=sched.go:466 msg="system memory" total="15.9 GiB" free="9.1 GiB" free_swap="16.7 GiB" time=2026-02-17T16:20:17.603+01:00 level=INFO source=sched.go:473 msg="gpu memory" id=GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 library=CUDA available="2.9 GiB" free="3.3 GiB" minimum="457.0 MiB" overhead="0 B" time=2026-02-17T16:20:17.603+01:00 level=INFO source=server.go:498 msg="loading model" "model layers"=49 requested=2 time=2026-02-17T16:20:17.604+01:00 level=INFO source=device.go:240 msg="model weights" device=CUDA0 size="332.8 MiB" time=2026-02-17T16:20:17.604+01:00 level=INFO source=device.go:245 msg="model weights" device=CPU size="7.6 GiB" time=2026-02-17T16:20:17.604+01:00 level=INFO source=device.go:251 msg="kv cache" device=CUDA0 size="32.0 MiB" time=2026-02-17T16:20:17.604+01:00 level=INFO source=device.go:256 msg="kv cache" device=CPU size="736.0 MiB" time=2026-02-17T16:20:17.604+01:00 level=INFO source=device.go:262 msg="compute graph" device=CUDA0 size="916.1 MiB" time=2026-02-17T16:20:17.604+01:00 level=INFO source=device.go:272 msg="total memory" size="9.6 GiB" time=2026-02-17T16:20:17.646+01:00 level=INFO source=runner.go:965 msg="starting go runner" load_backend: loaded CPU backend from C:\Users\robob\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no ggml_cuda_init: found 1 CUDA devices: Device 0: NVIDIA GeForce GTX 980, compute capability 5.2, VMM: yes, ID: GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 load_backend: loaded CUDA backend from C:\Users\robob\AppData\Local\Programs\Ollama\lib\ollama\cuda_v12\ggml-cuda.dll time=2026-02-17T16:20:17.835+01:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 CUDA.0.ARCHS=500,520,600,610,700,750,800,860,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(clang) time=2026-02-17T16:20:17.836+01:00 level=INFO source=runner.go:1001 msg="Server listening on 127.0.0.1:34062" time=2026-02-17T16:20:17.844+01:00 level=INFO source=runner.go:895 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Auto KvSize:4096 KvCacheType: NumThreads:4 GPULayers:2[ID:GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 Layers:2(46..47)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2026-02-17T16:20:17.844+01:00 level=INFO source=server.go:1350 msg="waiting for llama runner to start responding" time=2026-02-17T16:20:17.844+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server loading model" ggml_backend_cuda_device_get_memory device GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 utilizing NVML memory reporting free: 3549564928 total: 4294967296 llama_model_load_from_file_impl: using device CUDA0 (NVIDIA GeForce GTX 980) (0000:01:00.0) - 3385 MiB free llama_model_loader: loaded meta data with 26 key-value pairs and 579 tensors from F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e (version GGUF V3 (latest)) llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output. llama_model_loader: - kv 0: general.architecture str = qwen2 llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.name str = DeepSeek R1 Distill Qwen 14B llama_model_loader: - kv 3: general.basename str = DeepSeek-R1-Distill-Qwen llama_model_loader: - kv 4: general.size_label str = 14B llama_model_loader: - kv 5: qwen2.block_count u32 = 48 llama_model_loader: - kv 6: qwen2.context_length u32 = 131072 llama_model_loader: - kv 7: qwen2.embedding_length u32 = 5120 llama_model_loader: - kv 8: qwen2.feed_forward_length u32 = 13824 llama_model_loader: - kv 9: qwen2.attention.head_count u32 = 40 llama_model_loader: - kv 10: qwen2.attention.head_count_kv u32 = 8 llama_model_loader: - kv 11: qwen2.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 12: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 13: general.file_type u32 = 15 llama_model_loader: - kv 14: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 15: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 16: tokenizer.ggml.tokens arr[str,152064] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 18: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 151646 llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151643 llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 24: tokenizer.chat_template str = {% if not add_generation_prompt is de... llama_model_loader: - kv 25: general.quantization_version u32 = 2 llama_model_loader: - type f32: 241 tensors llama_model_loader: - type q4_K: 289 tensors llama_model_loader: - type q6_K: 49 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 8.37 GiB (4.87 BPW) load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect load: printing all EOG tokens: load: - 151643 ('<｜end▁of▁sentence｜>') load: - 151662 ('<|fim_pad|>') load: - 151663 ('<|repo_name|>') load: - 151664 ('<|file_sep|>') load: special tokens cache size = 22 load: token to piece cache size = 0.9310 MB print_info: arch = qwen2 print_info: vocab_only = 0 print_info: no_alloc = 0 print_info: n_ctx_train = 131072 print_info: n_embd = 5120 print_info: n_embd_inp = 5120 print_info: n_layer = 48 print_info: n_head = 40 print_info: n_head_kv = 8 print_info: n_rot = 128 print_info: n_swa = 0 print_info: is_swa_any = 0 print_info: n_embd_head_k = 128 print_info: n_embd_head_v = 128 print_info: n_gqa = 5 print_info: n_embd_k_gqa = 1024 print_info: n_embd_v_gqa = 1024 print_info: f_norm_eps = 0.0e+00 print_info: f_norm_rms_eps = 1.0e-05 print_info: f_clamp_kqv = 0.0e+00 print_info: f_max_alibi_bias = 0.0e+00 print_info: f_logit_scale = 0.0e+00 print_info: f_attn_scale = 0.0e+00 print_info: n_ff = 13824 print_info: n_expert = 0 print_info: n_expert_used = 0 print_info: n_expert_groups = 0 print_info: n_group_used = 0 print_info: causal attn = 1 print_info: pooling type = -1 print_info: rope type = 2 print_info: rope scaling = linear print_info: freq_base_train = 1000000.0 print_info: freq_scale_train = 1 print_info: n_ctx_orig_yarn = 131072 print_info: rope_yarn_log_mul= 0.0000 print_info: rope_finetuned = unknown print_info: model type = 14B print_info: model params = 14.77 B print_info: general.name = DeepSeek R1 Distill Qwen 14B print_info: vocab type = BPE print_info: n_vocab = 152064 print_info: n_merges = 151387 print_info: BOS token = 151646 '<｜begin▁of▁sentence｜>' print_info: EOS token = 151643 '<｜end▁of▁sentence｜>' print_info: EOT token = 151643 '<｜end▁of▁sentence｜>' print_info: PAD token = 151643 '<｜end▁of▁sentence｜>' print_info: LF token = 198 'Ċ' print_info: FIM PRE token = 151659 '<|fim_prefix|>' print_info: FIM SUF token = 151661 '<|fim_suffix|>' print_info: FIM MID token = 151660 '<|fim_middle|>' print_info: FIM PAD token = 151662 '<|fim_pad|>' print_info: FIM REP token = 151663 '<|repo_name|>' print_info: FIM SEP token = 151664 '<|file_sep|>' print_info: EOG token = 151643 '<｜end▁of▁sentence｜>' print_info: EOG token = 151662 '<|fim_pad|>' print_info: EOG token = 151663 '<|repo_name|>' print_info: EOG token = 151664 '<|file_sep|>' print_info: max token length = 256 load_tensors: loading model tensors, this can take a while... (mmap = false) ggml_cuda_host_malloc: failed to allocate 7815.55 MiB of pinned memory: out of memory load_tensors: offloading 2 repeating layers to GPU load_tensors: offloaded 2/49 layers to GPU load_tensors: CPU model buffer size = 417.66 MiB load_tensors: CUDA0 model buffer size = 332.83 MiB load_tensors: CPU model buffer size = 7815.55 MiB llama_context: constructing llama_context llama_context: n_seq_max = 1 llama_context: n_ctx = 4096 llama_context: n_ctx_seq = 4096 llama_context: n_batch = 512 llama_context: n_ubatch = 512 llama_context: causal_attn = 1 llama_context: flash_attn = auto llama_context: kv_unified = false llama_context: freq_base = 1000000.0 llama_context: freq_scale = 1 llama_context: n_ctx_seq (4096) < n_ctx_train (131072) -- the full capacity of the model will not be utilized llama_context: CPU output buffer size = 0.60 MiB llama_kv_cache: CPU KV buffer size = 736.00 MiB [GIN] 2026/02/17 - 16:20:41 | 200 | 14.9989ms | 127.0.0.1 | GET "/api/tags" llama_kv_cache: CUDA0 KV buffer size = 32.00 MiB llama_kv_cache: size = 768.00 MiB ( 4096 cells, 48 layers, 1/1 seqs), K (f16): 384.00 MiB, V (f16): 384.00 MiB llama_context: Flash Attention was auto, set to enabled graph_reserve: failed to allocate compute buffers Exception 0xc0000005 0x0 0x243fe0a9d60 0x7ffceb1fe53e PC=0x7ffceb1fe53e signal arrived during external code execution runtime.cgocall(0x7ff793e6e2b0, 0xc000365c00) runtime/cgocall.go:167 +0x3e fp=0xc000365bd8 sp=0xc000365b70 pc=0x7ff792fd243e github.com/ollama/ollama/llama._Cfunc_llama_init_from_model(0x258f4221790, {0x1000, 0x200, 0x200, 0x1, 0x4, 0x4, 0xffffffff, 0xffffffff, 0xffffffff, ...}) _cgo_gotypes.go:759 +0x54 fp=0xc000365c00 sp=0xc000365bd8 pc=0x7ff793479774 github.com/ollama/ollama/llama.NewContextWithModel.func1(...) github.com/ollama/ollama/llama/llama.go:322 github.com/ollama/ollama/llama.NewContextWithModel(0xc0001cc020, {{0x1000, 0x200, 0x200, 0x1, 0x4, 0x4, 0xffffffff, 0xffffffff, 0xffffffff, ...}}) github.com/ollama/ollama/llama/llama.go:322 +0x158 fp=0xc000365da0 sp=0xc000365c00 pc=0x7ff79347dd58 github.com/ollama/ollama/runner/llamarunner.(*Server).loadModel(0xc0003db0e0, {{0xc0000f58f8, 0x1, 0x1}, 0x2, 0x0, 0x0, {0xc0000f58f0, 0x1, 0x2}, ...}, ...) github.com/ollama/ollama/runner/llamarunner/runner.go:847 +0x178 fp=0xc000365ee8 sp=0xc000365da0 pc=0x7ff79352c938 github.com/ollama/ollama/runner/llamarunner.(*Server).load.gowrap2() github.com/ollama/ollama/runner/llamarunner/runner.go:934 +0x114 fp=0xc000365fe0 sp=0xc000365ee8 pc=0x7ff79352db54 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000365fe8 sp=0xc000365fe0 pc=0x7ff792fdd9a1 created by github.com/ollama/ollama/runner/llamarunner.(*Server).load in goroutine 12 github.com/ollama/ollama/runner/llamarunner/runner.go:934 +0x889 goroutine 1 gp=0xc0000021c0 m=nil [IO wait]: runtime.gopark(0x7ff792fdf1a0?, 0x7ff795180d00?, 0xa0?, 0xf1?, 0xc00047f24c?) runtime/proc.go:435 +0xce fp=0xc000587630 sp=0xc000587610 pc=0x7ff792fd598e runtime.netpollblock(0x380?, 0x92f70406?, 0xf7?) runtime/netpoll.go:575 +0xf7 fp=0xc000587668 sp=0xc000587630 pc=0x7ff792f9bdf7 internal/poll.runtime_pollWait(0x258f1616050, 0x72) runtime/netpoll.go:351 +0x85 fp=0xc000587688 sp=0xc000587668 pc=0x7ff792fd4b25 internal/poll.(*pollDesc).wait(0x7ff79306a8f3?, 0x0?, 0x0) internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc0005876b0 sp=0xc000587688 pc=0x7ff79306bee7 internal/poll.execIO(0xc00047f1a0, 0xc000587758) internal/poll/fd_windows.go:177 +0x105 fp=0xc000587728 sp=0xc0005876b0 pc=0x7ff79306d345 internal/poll.(*FD).acceptOne(0xc00047f188, 0x38c, {0xc0003560f0?, 0xc0005877b8?, 0x7ff793075005?}, 0xc0005877ec?) internal/poll/fd_windows.go:946 +0x65 fp=0xc000587788 sp=0xc000587728 pc=0x7ff7930718c5 internal/poll.(*FD).Accept(0xc00047f188, 0xc000587938) internal/poll/fd_windows.go:980 +0x1b6 fp=0xc000587840 sp=0xc000587788 pc=0x7ff793071bf6 net.(*netFD).accept(0xc00047f188) net/fd_windows.go:182 +0x4b fp=0xc000587958 sp=0xc000587840 pc=0x7ff7930e344b net.(*TCPListener).accept(0xc0002ae500) net/tcpsock_posix.go:159 +0x1b fp=0xc0005879a8 sp=0xc000587958 pc=0x7ff7930f99fb net.(*TCPListener).Accept(0xc0002ae500) net/tcpsock.go:380 +0x30 fp=0xc0005879d8 sp=0xc0005879a8 pc=0x7ff7930f87b0 net/http.(*onceCloseListener).Accept(0xc000204480?) <autogenerated>:1 +0x24 fp=0xc0005879f0 sp=0xc0005879d8 pc=0x7ff793311ec4 net/http.(*Server).Serve(0xc00013fd00, {0x7ff7946e21d0, 0xc0002ae500}) net/http/server.go:3424 +0x30c fp=0xc000587b20 sp=0xc0005879f0 pc=0x7ff7932e978c github.com/ollama/ollama/runner/llamarunner.Execute({0xc0000a4020, 0x4, 0x6}) github.com/ollama/ollama/runner/llamarunner/runner.go:1002 +0x8f5 fp=0xc000587cf0 sp=0xc000587b20 pc=0x7ff79352e515 github.com/ollama/ollama/runner.Execute({0xc0000a4010?, 0x0?, 0x0?}) github.com/ollama/ollama/runner/runner.go:25 +0x1a5 fp=0xc000587d30 sp=0xc000587cf0 pc=0x7ff793603ae5 github.com/ollama/ollama/cmd.NewCLI.func3(0xc00013f800?, {0x7ff7944c4587?, 0x4?, 0x7ff7944c458b?}) github.com/ollama/ollama/cmd/cmd.go:2262 +0x45 fp=0xc000587d58 sp=0xc000587d30 pc=0x7ff793dfce65 github.com/spf13/cobra.(*Command).execute(0xc0002d3b08, {0xc0002ae280, 0x4, 0x4}) github.com/spf13/cobra@v1.7.0/command.go:940 +0x85c fp=0xc000587e78 sp=0xc000587d58 pc=0x7ff79315e61c github.com/spf13/cobra.(*Command).ExecuteC(0xc0004d6f08) github.com/spf13/cobra@v1.7.0/command.go:1068 +0x3a5 fp=0xc000587f30 sp=0xc000587e78 pc=0x7ff79315ee65 github.com/spf13/cobra.(*Command).Execute(...) github.com/spf13/cobra@v1.7.0/command.go:992 github.com/spf13/cobra.(*Command).ExecuteContext(...) github.com/spf13/cobra@v1.7.0/command.go:985 main.main() github.com/ollama/ollama/main.go:12 +0x4d fp=0xc000587f50 sp=0xc000587f30 pc=0x7ff793dff28d runtime.main() runtime/proc.go:283 +0x27d fp=0xc000587fe0 sp=0xc000587f50 pc=0x7ff792fa4ddd runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000587fe8 sp=0xc000587fe0 pc=0x7ff792fdd9a1 goroutine 2 gp=0xc0000028c0 m=nil [force gc (idle)]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000071fa8 sp=0xc000071f88 pc=0x7ff792fd598e runtime.goparkunlock(...) runtime/proc.go:441 runtime.forcegchelper() runtime/proc.go:348 +0xb8 fp=0xc000071fe0 sp=0xc000071fa8 pc=0x7ff792fa50f8 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000071fe8 sp=0xc000071fe0 pc=0x7ff792fdd9a1 created by runtime.init.7 in goroutine 1 runtime/proc.go:336 +0x1a goroutine 3 gp=0xc000002c40 m=nil [GC sweep wait]: runtime.gopark(0x1?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000073f80 sp=0xc000073f60 pc=0x7ff792fd598e runtime.goparkunlock(...) runtime/proc.go:441 runtime.bgsweep(0xc000080000) runtime/mgcsweep.go:316 +0xdf fp=0xc000073fc8 sp=0xc000073f80 pc=0x7ff792f8debf runtime.gcenable.gowrap1() runtime/mgc.go:204 +0x25 fp=0xc000073fe0 sp=0xc000073fc8 pc=0x7ff792f82285 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000073fe8 sp=0xc000073fe0 pc=0x7ff792fdd9a1 created by runtime.gcenable in goroutine 1 runtime/mgc.go:204 +0x66 goroutine 4 gp=0xc000002e00 m=nil [GC scavenge wait]: runtime.gopark(0x10000?, 0x7ff7946ccc38?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000087f78 sp=0xc000087f58 pc=0x7ff792fd598e runtime.goparkunlock(...) runtime/proc.go:441 runtime.(*scavengerState).park(0x7ff7951aa900) runtime/mgcscavenge.go:425 +0x49 fp=0xc000087fa8 sp=0xc000087f78 pc=0x7ff792f8b909 runtime.bgscavenge(0xc000080000) runtime/mgcscavenge.go:658 +0x59 fp=0xc000087fc8 sp=0xc000087fa8 pc=0x7ff792f8be99 runtime.gcenable.gowrap2() runtime/mgc.go:205 +0x25 fp=0xc000087fe0 sp=0xc000087fc8 pc=0x7ff792f82225 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000087fe8 sp=0xc000087fe0 pc=0x7ff792fdd9a1 created by runtime.gcenable in goroutine 1 runtime/mgc.go:205 +0xa5 goroutine 5 gp=0xc000003340 m=nil [finalizer wait]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000089e30 sp=0xc000089e10 pc=0x7ff792fd598e runtime.runfinq() runtime/mfinal.go:196 +0x107 fp=0xc000089fe0 sp=0xc000089e30 pc=0x7ff792f81207 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000089fe8 sp=0xc000089fe0 pc=0x7ff792fdd9a1 created by runtime.createfing in goroutine 1 runtime/mfinal.go:166 +0x3d goroutine 6 gp=0xc000003dc0 m=nil [chan receive]: runtime.gopark(0xc00017b2c0?, 0xc000508018?, 0x60?, 0x5f?, 0x7ff7930cc148?) runtime/proc.go:435 +0xce fp=0xc000075f18 sp=0xc000075ef8 pc=0x7ff792fd598e runtime.chanrecv(0xc00008c380, 0x0, 0x1) runtime/chan.go:664 +0x445 fp=0xc000075f90 sp=0xc000075f18 pc=0x7ff792f72d45 runtime.chanrecv1(0x7ff792fa4f40?, 0xc000075f76?) runtime/chan.go:506 +0x12 fp=0xc000075fb8 sp=0xc000075f90 pc=0x7ff792f728d2 runtime.unique_runtime_registerUniqueMapCleanup.func2(...) runtime/mgc.go:1796 runtime.unique_runtime_registerUniqueMapCleanup.gowrap1() runtime/mgc.go:1799 +0x2f fp=0xc000075fe0 sp=0xc000075fb8 pc=0x7ff792f854af runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000075fe8 sp=0xc000075fe0 pc=0x7ff792fdd9a1 created by unique.runtime_registerUniqueMapCleanup in goroutine 1 runtime/mgc.go:1794 +0x85 goroutine 7 gp=0xc0003d8540 m=nil [GC worker (idle)]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000083f38 sp=0xc000083f18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008d960) runtime/mgc.go:1423 +0xe9 fp=0xc000083fc8 sp=0xc000083f38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc000083fe0 sp=0xc000083fc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000083fe8 sp=0xc000083fe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 8 gp=0xc0003d8700 m=nil [GC worker (idle)]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000085f38 sp=0xc000085f18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008d960) runtime/mgc.go:1423 +0xe9 fp=0xc000085fc8 sp=0xc000085f38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc000085fe0 sp=0xc000085fc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000085fe8 sp=0xc000085fe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 9 gp=0xc0003d88c0 m=nil [GC worker (idle)]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc00047bf38 sp=0xc00047bf18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008d960) runtime/mgc.go:1423 +0xe9 fp=0xc00047bfc8 sp=0xc00047bf38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc00047bfe0 sp=0xc00047bfc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc00047bfe8 sp=0xc00047bfe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 18 gp=0xc0004861c0 m=nil [GC worker (idle)]: runtime.gopark(0x4bd0e233d4d0?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000477f38 sp=0xc000477f18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008d960) runtime/mgc.go:1423 +0xe9 fp=0xc000477fc8 sp=0xc000477f38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc000477fe0 sp=0xc000477fc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000477fe8 sp=0xc000477fe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 19 gp=0xc000486380 m=nil [GC worker (idle)]: runtime.gopark(0x4bd0e233d4d0?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000479f38 sp=0xc000479f18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008d960) runtime/mgc.go:1423 +0xe9 fp=0xc000479fc8 sp=0xc000479f38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc000479fe0 sp=0xc000479fc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000479fe8 sp=0xc000479fe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 20 gp=0xc000486540 m=nil [GC worker (idle)]: runtime.gopark(0x4bd0e233d4d0?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000491f38 sp=0xc000491f18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008d960) runtime/mgc.go:1423 +0xe9 fp=0xc000491fc8 sp=0xc000491f38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc000491fe0 sp=0xc000491fc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000491fe8 sp=0xc000491fe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 34 gp=0xc000206000 m=nil [GC worker (idle)]: runtime.gopark(0x4bd0e233d4d0?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc00048df38 sp=0xc00048df18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008d960) runtime/mgc.go:1423 +0xe9 fp=0xc00048dfc8 sp=0xc00048df38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc00048dfe0 sp=0xc00048dfc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc00048dfe8 sp=0xc00048dfe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 10 gp=0xc0003d8a80 m=nil [GC worker (idle)]: runtime.gopark(0x4bd0e233d4d0?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc00047df38 sp=0xc00047df18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008d960) runtime/mgc.go:1423 +0xe9 fp=0xc00047dfc8 sp=0xc00047df38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc00047dfe0 sp=0xc00047dfc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc00047dfe8 sp=0xc00047dfe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 11 gp=0xc0002061c0 m=nil [sync.WaitGroup.Wait]: runtime.gopark(0x0?, 0x0?, 0x20?, 0x1?, 0x0?) runtime/proc.go:435 +0xce fp=0xc00048fe20 sp=0xc00048fe00 pc=0x7ff792fd598e runtime.goparkunlock(...) runtime/proc.go:441 runtime.semacquire1(0xc0003db100, 0x0, 0x1, 0x0, 0x18) runtime/sema.go:188 +0x22f fp=0xc00048fe88 sp=0xc00048fe20 pc=0x7ff792fb750f sync.runtime_SemacquireWaitGroup(0x0?) runtime/sema.go:110 +0x25 fp=0xc00048fec0 sp=0xc00048fe88 pc=0x7ff792fd6f85 sync.(*WaitGroup).Wait(0x0?) sync/waitgroup.go:118 +0x48 fp=0xc00048fee8 sp=0xc00048fec0 pc=0x7ff792feb988 github.com/ollama/ollama/runner/llamarunner.(*Server).run(0xc0003db0e0, {0x7ff7946e4a30, 0xc00033e550}) github.com/ollama/ollama/runner/llamarunner/runner.go:360 +0x4b fp=0xc00048ffb8 sp=0xc00048fee8 pc=0x7ff79352922b github.com/ollama/ollama/runner/llamarunner.Execute.gowrap1() github.com/ollama/ollama/runner/llamarunner/runner.go:981 +0x28 fp=0xc00048ffe0 sp=0xc00048ffb8 pc=0x7ff79352e788 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc00048ffe8 sp=0xc00048ffe0 pc=0x7ff792fdd9a1 created by github.com/ollama/ollama/runner/llamarunner.Execute in goroutine 1 github.com/ollama/ollama/runner/llamarunner/runner.go:981 +0x4c5 goroutine 12 gp=0xc000206380 m=nil [IO wait]: runtime.gopark(0x0?, 0xc00047f420?, 0xc8?, 0xf4?, 0xc00047f4cc?) runtime/proc.go:435 +0xce fp=0xc00004b8c8 sp=0xc00004b8a8 pc=0x7ff792fd598e runtime.netpollblock(0x388?, 0x92f70406?, 0xf7?) runtime/netpoll.go:575 +0xf7 fp=0xc00004b900 sp=0xc00004b8c8 pc=0x7ff792f9bdf7 internal/poll.runtime_pollWait(0x258f1615f38, 0x72) runtime/netpoll.go:351 +0x85 fp=0xc00004b920 sp=0xc00004b900 pc=0x7ff792fd4b25 internal/poll.(*pollDesc).wait(0x7ff79319d5f7?, 0xc00004b970?, 0x0) internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc00004b948 sp=0xc00004b920 pc=0x7ff79306bee7 internal/poll.execIO(0xc00047f420, 0x7ff794546098) internal/poll/fd_windows.go:177 +0x105 fp=0xc00004b9c0 sp=0xc00004b948 pc=0x7ff79306d345 internal/poll.(*FD).Read(0xc00047f408, {0xc000091000, 0x1000, 0x1000}) internal/poll/fd_windows.go:438 +0x29b fp=0xc00004ba60 sp=0xc00004b9c0 pc=0x7ff79306e01b net.(*netFD).Read(0xc00047f408, {0xc000091000?, 0xc00004bad0?, 0x7ff79306c3a5?}) net/fd_posix.go:55 +0x25 fp=0xc00004baa8 sp=0xc00004ba60 pc=0x7ff7930e1325 net.(*conn).Read(0xc0004ac530, {0xc000091000?, 0x0?, 0x0?}) net/net.go:194 +0x45 fp=0xc00004baf0 sp=0xc00004baa8 pc=0x7ff7930f0a45 net/http.(*connReader).Read(0xc0003242d0, {0xc000091000, 0x1000, 0x1000}) net/http/server.go:798 +0x159 fp=0xc00004bb40 sp=0xc00004baf0 pc=0x7ff7932de639 bufio.(*Reader).fill(0xc000200660) bufio/bufio.go:113 +0x103 fp=0xc00004bb78 sp=0xc00004bb40 pc=0x7ff793107963 bufio.(*Reader).Peek(0xc000200660, 0x4) bufio/bufio.go:152 +0x53 fp=0xc00004bb98 sp=0xc00004bb78 pc=0x7ff793107a93 net/http.(*conn).serve(0xc000204480, {0x7ff7946e49f8, 0xc000324000}) net/http/server.go:2137 +0x785 fp=0xc00004bfb8 sp=0xc00004bb98 pc=0x7ff7932e4425 net/http.(*Server).Serve.gowrap3() net/http/server.go:3454 +0x28 fp=0xc00004bfe0 sp=0xc00004bfb8 pc=0x7ff7932e9b88 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc00004bfe8 sp=0xc00004bfe0 pc=0x7ff792fdd9a1 created by net/http.(*Server).Serve in goroutine 1 net/http/server.go:3454 +0x485 rax 0xffffffeb02005900 rbx 0x258fc0a4460 rcx 0x258f4258d40 rdx 0x0 rdi 0x0 rsi 0x258f42b3230 rbp 0xf997aff9c0 rsp 0xf997aff710 r8 0x258ac180f00 r9 0x1 r10 0x8000 r11 0xf997aff720 r12 0x667 r13 0x1218 r14 0x1 r15 0x258f4221790 rip 0x7ffceb1fe53e rflags 0x10287 cs 0x33 fs 0x53 gs 0x2b time=2026-02-17T16:20:43.383+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server not responding" time=2026-02-17T16:20:43.634+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server error" time=2026-02-17T16:20:44.386+01:00 level=INFO source=sched.go:493 msg="Load failed" model=F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e error="llama runner process has terminated: exit status 2" [GIN] 2026/02/17 - 16:20:44 | 500 | 27.5168873s | 127.0.0.1 | POST "/api/chat" [GIN] 2026/02/17 - 16:20:49 | 200 | 134.789ms | 127.0.0.1 | POST "/api/show" [GIN] 2026/02/17 - 16:20:56 | 200 | 8.8411ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:21:00 | 200 | 11.267ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:21:00 | 200 | 135.358ms | 127.0.0.1 | POST "/api/show" [GIN] 2026/02/17 - 16:21:01 | 200 | 119.4627ms | 127.0.0.1 | POST "/api/show" time=2026-02-17T16:21:01.197+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 34095" time=2026-02-17T16:21:01.673+01:00 level=INFO source=cpu_windows.go:148 msg=packages count=1 time=2026-02-17T16:21:01.673+01:00 level=INFO source=cpu_windows.go:195 msg="" package=0 cores=4 efficiency=0 threads=8 llama_model_loader: loaded meta data with 26 key-value pairs and 579 tensors from F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e (version GGUF V3 (latest)) llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output. llama_model_loader: - kv 0: general.architecture str = qwen2 llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.name str = DeepSeek R1 Distill Qwen 14B llama_model_loader: - kv 3: general.basename str = DeepSeek-R1-Distill-Qwen llama_model_loader: - kv 4: general.size_label str = 14B llama_model_loader: - kv 5: qwen2.block_count u32 = 48 llama_model_loader: - kv 6: qwen2.context_length u32 = 131072 llama_model_loader: - kv 7: qwen2.embedding_length u32 = 5120 llama_model_loader: - kv 8: qwen2.feed_forward_length u32 = 13824 llama_model_loader: - kv 9: qwen2.attention.head_count u32 = 40 llama_model_loader: - kv 10: qwen2.attention.head_count_kv u32 = 8 llama_model_loader: - kv 11: qwen2.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 12: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 13: general.file_type u32 = 15 llama_model_loader: - kv 14: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 15: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 16: tokenizer.ggml.tokens arr[str,152064] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 18: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 151646 llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151643 llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 24: tokenizer.chat_template str = {% if not add_generation_prompt is de... llama_model_loader: - kv 25: general.quantization_version u32 = 2 llama_model_loader: - type f32: 241 tensors llama_model_loader: - type q4_K: 289 tensors llama_model_loader: - type q6_K: 49 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 8.37 GiB (4.87 BPW) load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect load: printing all EOG tokens: load: - 151643 ('<｜end▁of▁sentence｜>') load: - 151662 ('<|fim_pad|>') load: - 151663 ('<|repo_name|>') load: - 151664 ('<|file_sep|>') load: special tokens cache size = 22 load: token to piece cache size = 0.9310 MB print_info: arch = qwen2 print_info: vocab_only = 1 print_info: no_alloc = 0 print_info: model type = ?B print_info: model params = 14.77 B print_info: general.name = DeepSeek R1 Distill Qwen 14B print_info: vocab type = BPE print_info: n_vocab = 152064 print_info: n_merges = 151387 print_info: BOS token = 151646 '<｜begin▁of▁sentence｜>' print_info: EOS token = 151643 '<｜end▁of▁sentence｜>' print_info: EOT token = 151643 '<｜end▁of▁sentence｜>' print_info: PAD token = 151643 '<｜end▁of▁sentence｜>' print_info: LF token = 198 'Ċ' print_info: FIM PRE token = 151659 '<|fim_prefix|>' print_info: FIM SUF token = 151661 '<|fim_suffix|>' print_info: FIM MID token = 151660 '<|fim_middle|>' print_info: FIM PAD token = 151662 '<|fim_pad|>' print_info: FIM REP token = 151663 '<|repo_name|>' print_info: FIM SEP token = 151664 '<|file_sep|>' print_info: EOG token = 151643 '<｜end▁of▁sentence｜>' print_info: EOG token = 151662 '<|fim_pad|>' print_info: EOG token = 151663 '<|repo_name|>' print_info: EOG token = 151664 '<|file_sep|>' print_info: max token length = 256 llama_model_load: vocab only - skipping tensors time=2026-02-17T16:21:02.012+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --model F:\\OllamaModels\\blobs\\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e --port 34125" time=2026-02-17T16:21:02.013+01:00 level=INFO source=sched.go:466 msg="system memory" total="15.9 GiB" free="9.3 GiB" free_swap="16.8 GiB" time=2026-02-17T16:21:02.013+01:00 level=INFO source=sched.go:473 msg="gpu memory" id=GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 library=CUDA available="3.1 GiB" free="3.6 GiB" minimum="457.0 MiB" overhead="0 B" time=2026-02-17T16:21:02.014+01:00 level=INFO source=server.go:498 msg="loading model" "model layers"=49 requested=1 time=2026-02-17T16:21:02.014+01:00 level=INFO source=device.go:240 msg="model weights" device=CUDA0 size="166.4 MiB" time=2026-02-17T16:21:02.014+01:00 level=INFO source=device.go:245 msg="model weights" device=CPU size="7.8 GiB" time=2026-02-17T16:21:02.014+01:00 level=INFO source=device.go:251 msg="kv cache" device=CUDA0 size="16.0 MiB" time=2026-02-17T16:21:02.014+01:00 level=INFO source=device.go:256 msg="kv cache" device=CPU size="752.0 MiB" time=2026-02-17T16:21:02.014+01:00 level=INFO source=device.go:262 msg="compute graph" device=CUDA0 size="916.1 MiB" time=2026-02-17T16:21:02.014+01:00 level=INFO source=device.go:272 msg="total memory" size="9.6 GiB" time=2026-02-17T16:21:02.058+01:00 level=INFO source=runner.go:965 msg="starting go runner" load_backend: loaded CPU backend from C:\Users\robob\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no ggml_cuda_init: found 1 CUDA devices: Device 0: NVIDIA GeForce GTX 980, compute capability 5.2, VMM: yes, ID: GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 load_backend: loaded CUDA backend from C:\Users\robob\AppData\Local\Programs\Ollama\lib\ollama\cuda_v12\ggml-cuda.dll time=2026-02-17T16:21:02.251+01:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 CUDA.0.ARCHS=500,520,600,610,700,750,800,860,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(clang) time=2026-02-17T16:21:02.252+01:00 level=INFO source=runner.go:1001 msg="Server listening on 127.0.0.1:34125" time=2026-02-17T16:21:02.255+01:00 level=INFO source=runner.go:895 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Auto KvSize:4096 KvCacheType: NumThreads:4 GPULayers:1[ID:GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 Layers:1(47..47)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2026-02-17T16:21:02.255+01:00 level=INFO source=server.go:1350 msg="waiting for llama runner to start responding" time=2026-02-17T16:21:02.255+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server loading model" ggml_backend_cuda_device_get_memory device GPU-03c39fd0-a4d6-cb45-7283-3e477c7c0cd2 utilizing NVML memory reporting free: 3815043072 total: 4294967296 llama_model_load_from_file_impl: using device CUDA0 (NVIDIA GeForce GTX 980) (0000:01:00.0) - 3638 MiB free llama_model_loader: loaded meta data with 26 key-value pairs and 579 tensors from F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e (version GGUF V3 (latest)) llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output. llama_model_loader: - kv 0: general.architecture str = qwen2 llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.name str = DeepSeek R1 Distill Qwen 14B llama_model_loader: - kv 3: general.basename str = DeepSeek-R1-Distill-Qwen llama_model_loader: - kv 4: general.size_label str = 14B llama_model_loader: - kv 5: qwen2.block_count u32 = 48 llama_model_loader: - kv 6: qwen2.context_length u32 = 131072 llama_model_loader: - kv 7: qwen2.embedding_length u32 = 5120 llama_model_loader: - kv 8: qwen2.feed_forward_length u32 = 13824 llama_model_loader: - kv 9: qwen2.attention.head_count u32 = 40 llama_model_loader: - kv 10: qwen2.attention.head_count_kv u32 = 8 llama_model_loader: - kv 11: qwen2.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 12: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 13: general.file_type u32 = 15 llama_model_loader: - kv 14: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 15: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 16: tokenizer.ggml.tokens arr[str,152064] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 18: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 151646 llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151643 llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 24: tokenizer.chat_template str = {% if not add_generation_prompt is de... llama_model_loader: - kv 25: general.quantization_version u32 = 2 llama_model_loader: - type f32: 241 tensors llama_model_loader: - type q4_K: 289 tensors llama_model_loader: - type q6_K: 49 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 8.37 GiB (4.87 BPW) load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect load: printing all EOG tokens: load: - 151643 ('<｜end▁of▁sentence｜>') load: - 151662 ('<|fim_pad|>') load: - 151663 ('<|repo_name|>') load: - 151664 ('<|file_sep|>') load: special tokens cache size = 22 load: token to piece cache size = 0.9310 MB print_info: arch = qwen2 print_info: vocab_only = 0 print_info: no_alloc = 0 print_info: n_ctx_train = 131072 print_info: n_embd = 5120 print_info: n_embd_inp = 5120 print_info: n_layer = 48 print_info: n_head = 40 print_info: n_head_kv = 8 print_info: n_rot = 128 print_info: n_swa = 0 print_info: is_swa_any = 0 print_info: n_embd_head_k = 128 print_info: n_embd_head_v = 128 print_info: n_gqa = 5 print_info: n_embd_k_gqa = 1024 print_info: n_embd_v_gqa = 1024 print_info: f_norm_eps = 0.0e+00 print_info: f_norm_rms_eps = 1.0e-05 print_info: f_clamp_kqv = 0.0e+00 print_info: f_max_alibi_bias = 0.0e+00 print_info: f_logit_scale = 0.0e+00 print_info: f_attn_scale = 0.0e+00 print_info: n_ff = 13824 print_info: n_expert = 0 print_info: n_expert_used = 0 print_info: n_expert_groups = 0 print_info: n_group_used = 0 print_info: causal attn = 1 print_info: pooling type = -1 print_info: rope type = 2 print_info: rope scaling = linear print_info: freq_base_train = 1000000.0 print_info: freq_scale_train = 1 print_info: n_ctx_orig_yarn = 131072 print_info: rope_yarn_log_mul= 0.0000 print_info: rope_finetuned = unknown print_info: model type = 14B print_info: model params = 14.77 B print_info: general.name = DeepSeek R1 Distill Qwen 14B print_info: vocab type = BPE print_info: n_vocab = 152064 print_info: n_merges = 151387 print_info: BOS token = 151646 '<｜begin▁of▁sentence｜>' print_info: EOS token = 151643 '<｜end▁of▁sentence｜>' print_info: EOT token = 151643 '<｜end▁of▁sentence｜>' print_info: PAD token = 151643 '<｜end▁of▁sentence｜>' print_info: LF token = 198 'Ċ' print_info: FIM PRE token = 151659 '<|fim_prefix|>' print_info: FIM SUF token = 151661 '<|fim_suffix|>' print_info: FIM MID token = 151660 '<|fim_middle|>' print_info: FIM PAD token = 151662 '<|fim_pad|>' print_info: FIM REP token = 151663 '<|repo_name|>' print_info: FIM SEP token = 151664 '<|file_sep|>' print_info: EOG token = 151643 '<｜end▁of▁sentence｜>' print_info: EOG token = 151662 '<|fim_pad|>' print_info: EOG token = 151663 '<|repo_name|>' print_info: EOG token = 151664 '<|file_sep|>' print_info: max token length = 256 load_tensors: loading model tensors, this can take a while... (mmap = false) ggml_cuda_host_malloc: failed to allocate 7981.97 MiB of pinned memory: out of memory load_tensors: offloading 1 repeating layers to GPU load_tensors: offloaded 1/49 layers to GPU load_tensors: CPU model buffer size = 417.66 MiB load_tensors: CUDA0 model buffer size = 166.41 MiB load_tensors: CPU model buffer size = 7981.97 MiB llama_context: constructing llama_context llama_context: n_seq_max = 1 llama_context: n_ctx = 4096 llama_context: n_ctx_seq = 4096 llama_context: n_batch = 512 llama_context: n_ubatch = 512 llama_context: causal_attn = 1 llama_context: flash_attn = auto llama_context: kv_unified = false llama_context: freq_base = 1000000.0 llama_context: freq_scale = 1 llama_context: n_ctx_seq (4096) < n_ctx_train (131072) -- the full capacity of the model will not be utilized llama_context: CPU output buffer size = 0.60 MiB llama_kv_cache: CPU KV buffer size = 752.00 MiB llama_kv_cache: CUDA0 KV buffer size = 16.00 MiB llama_kv_cache: size = 768.00 MiB ( 4096 cells, 48 layers, 1/1 seqs), K (f16): 384.00 MiB, V (f16): 384.00 MiB llama_context: Flash Attention was auto, set to enabled graph_reserve: failed to allocate compute buffers Exception 0xc0000005 0x0 0x1d7cc1fea10 0x7ffceb1fe53e PC=0x7ffceb1fe53e signal arrived during external code execution runtime.cgocall(0x7ff793e6e2b0, 0xc00022fc00) runtime/cgocall.go:167 +0x3e fp=0xc00022fbd8 sp=0xc00022fb70 pc=0x7ff792fd243e github.com/ollama/ollama/llama._Cfunc_llama_init_from_model(0x1eafd0e00b0, {0x1000, 0x200, 0x200, 0x1, 0x4, 0x4, 0xffffffff, 0xffffffff, 0xffffffff, ...}) _cgo_gotypes.go:759 +0x54 fp=0xc00022fc00 sp=0xc00022fbd8 pc=0x7ff793479774 github.com/ollama/ollama/llama.NewContextWithModel.func1(...) github.com/ollama/ollama/llama/llama.go:322 github.com/ollama/ollama/llama.NewContextWithModel(0xc0001cd1e0, {{0x1000, 0x200, 0x200, 0x1, 0x4, 0x4, 0xffffffff, 0xffffffff, 0xffffffff, ...}}) github.com/ollama/ollama/llama/llama.go:322 +0x158 fp=0xc00022fda0 sp=0xc00022fc00 pc=0x7ff79347dd58 github.com/ollama/ollama/runner/llamarunner.(*Server).loadModel(0xc00043fae0, {{0xc0003806f0, 0x1, 0x1}, 0x1, 0x0, 0x0, {0xc0003806e8, 0x1, 0x2}, ...}, ...) github.com/ollama/ollama/runner/llamarunner/runner.go:847 +0x178 fp=0xc00022fee8 sp=0xc00022fda0 pc=0x7ff79352c938 github.com/ollama/ollama/runner/llamarunner.(*Server).load.gowrap2() github.com/ollama/ollama/runner/llamarunner/runner.go:934 +0x114 fp=0xc00022ffe0 sp=0xc00022fee8 pc=0x7ff79352db54 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc00022ffe8 sp=0xc00022ffe0 pc=0x7ff792fdd9a1 created by github.com/ollama/ollama/runner/llamarunner.(*Server).load in goroutine 11 github.com/ollama/ollama/runner/llamarunner/runner.go:934 +0x889 goroutine 1 gp=0xc0000021c0 m=nil [IO wait]: runtime.gopark(0x7ff792fdf1a0?, 0x7ff795180d00?, 0x20?, 0xd4?, 0xc00060d4cc?) runtime/proc.go:435 +0xce fp=0xc0004c9630 sp=0xc0004c9610 pc=0x7ff792fd598e runtime.netpollblock(0x384?, 0x92f70406?, 0xf7?) runtime/netpoll.go:575 +0xf7 fp=0xc0004c9668 sp=0xc0004c9630 pc=0x7ff792f9bdf7 internal/poll.runtime_pollWait(0x1eafa29c470, 0x72) runtime/netpoll.go:351 +0x85 fp=0xc0004c9688 sp=0xc0004c9668 pc=0x7ff792fd4b25 internal/poll.(*pollDesc).wait(0x7ff79306a8f3?, 0x0?, 0x0) internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc0004c96b0 sp=0xc0004c9688 pc=0x7ff79306bee7 internal/poll.execIO(0xc00060d420, 0xc00048d758) internal/poll/fd_windows.go:177 +0x105 fp=0xc0004c9728 sp=0xc0004c96b0 pc=0x7ff79306d345 internal/poll.(*FD).acceptOne(0xc00060d408, 0x390, {0xc0006500f0?, 0xc00048d7b8?, 0x7ff793075005?}, 0xc00048d7ec?) internal/poll/fd_windows.go:946 +0x65 fp=0xc0004c9788 sp=0xc0004c9728 pc=0x7ff7930718c5 internal/poll.(*FD).Accept(0xc00060d408, 0xc0004c9938) internal/poll/fd_windows.go:980 +0x1b6 fp=0xc0004c9840 sp=0xc0004c9788 pc=0x7ff793071bf6 net.(*netFD).accept(0xc00060d408) net/fd_windows.go:182 +0x4b fp=0xc0004c9958 sp=0xc0004c9840 pc=0x7ff7930e344b net.(*TCPListener).accept(0xc0003d46c0) net/tcpsock_posix.go:159 +0x1b fp=0xc0004c99a8 sp=0xc0004c9958 pc=0x7ff7930f99fb net.(*TCPListener).Accept(0xc0003d46c0) net/tcpsock.go:380 +0x30 fp=0xc0004c99d8 sp=0xc0004c99a8 pc=0x7ff7930f87b0 net/http.(*onceCloseListener).Accept(0xc000632480?) <autogenerated>:1 +0x24 fp=0xc0004c99f0 sp=0xc0004c99d8 pc=0x7ff793311ec4 net/http.(*Server).Serve(0xc000141d00, {0x7ff7946e21d0, 0xc0003d46c0}) net/http/server.go:3424 +0x30c fp=0xc0004c9b20 sp=0xc0004c99f0 pc=0x7ff7932e978c github.com/ollama/ollama/runner/llamarunner.Execute({0xc0000a4020, 0x4, 0x6}) github.com/ollama/ollama/runner/llamarunner/runner.go:1002 +0x8f5 fp=0xc0004c9cf0 sp=0xc0004c9b20 pc=0x7ff79352e515 github.com/ollama/ollama/runner.Execute({0xc0000a4010?, 0x0?, 0x0?}) github.com/ollama/ollama/runner/runner.go:25 +0x1a5 fp=0xc0004c9d30 sp=0xc0004c9cf0 pc=0x7ff793603ae5 github.com/ollama/ollama/cmd.NewCLI.func3(0xc000141800?, {0x7ff7944c4587?, 0x4?, 0x7ff7944c458b?}) github.com/ollama/ollama/cmd/cmd.go:2262 +0x45 fp=0xc0004c9d58 sp=0xc0004c9d30 pc=0x7ff793dfce65 github.com/spf13/cobra.(*Command).execute(0xc000149508, {0xc0003d4440, 0x4, 0x4}) github.com/spf13/cobra@v1.7.0/command.go:940 +0x85c fp=0xc0004c9e78 sp=0xc0004c9d58 pc=0x7ff79315e61c github.com/spf13/cobra.(*Command).ExecuteC(0xc000714c08) github.com/spf13/cobra@v1.7.0/command.go:1068 +0x3a5 fp=0xc0004c9f30 sp=0xc0004c9e78 pc=0x7ff79315ee65 github.com/spf13/cobra.(*Command).Execute(...) github.com/spf13/cobra@v1.7.0/command.go:992 github.com/spf13/cobra.(*Command).ExecuteContext(...) github.com/spf13/cobra@v1.7.0/command.go:985 main.main() github.com/ollama/ollama/main.go:12 +0x4d fp=0xc0004c9f50 sp=0xc0004c9f30 pc=0x7ff793dff28d runtime.main() runtime/proc.go:283 +0x27d fp=0xc0004c9fe0 sp=0xc0004c9f50 pc=0x7ff792fa4ddd runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc0004c9fe8 sp=0xc0004c9fe0 pc=0x7ff792fdd9a1 goroutine 2 gp=0xc0000028c0 m=nil [force gc (idle)]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000071fa8 sp=0xc000071f88 pc=0x7ff792fd598e runtime.goparkunlock(...) runtime/proc.go:441 runtime.forcegchelper() runtime/proc.go:348 +0xb8 fp=0xc000071fe0 sp=0xc000071fa8 pc=0x7ff792fa50f8 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000071fe8 sp=0xc000071fe0 pc=0x7ff792fdd9a1 created by runtime.init.7 in goroutine 1 runtime/proc.go:336 +0x1a goroutine 3 gp=0xc000002c40 m=nil [GC sweep wait]: runtime.gopark(0x1?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000073f80 sp=0xc000073f60 pc=0x7ff792fd598e runtime.goparkunlock(...) runtime/proc.go:441 runtime.bgsweep(0xc000080000) runtime/mgcsweep.go:316 +0xdf fp=0xc000073fc8 sp=0xc000073f80 pc=0x7ff792f8debf runtime.gcenable.gowrap1() runtime/mgc.go:204 +0x25 fp=0xc000073fe0 sp=0xc000073fc8 pc=0x7ff792f82285 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000073fe8 sp=0xc000073fe0 pc=0x7ff792fdd9a1 created by runtime.gcenable in goroutine 1 runtime/mgc.go:204 +0x66 goroutine 4 gp=0xc000002e00 m=nil [GC scavenge wait]: runtime.gopark(0x10000?, 0x7ff7946ccc38?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000087f78 sp=0xc000087f58 pc=0x7ff792fd598e runtime.goparkunlock(...) runtime/proc.go:441 runtime.(*scavengerState).park(0x7ff7951aa900) runtime/mgcscavenge.go:425 +0x49 fp=0xc000087fa8 sp=0xc000087f78 pc=0x7ff792f8b909 runtime.bgscavenge(0xc000080000) runtime/mgcscavenge.go:658 +0x59 fp=0xc000087fc8 sp=0xc000087fa8 pc=0x7ff792f8be99 runtime.gcenable.gowrap2() runtime/mgc.go:205 +0x25 fp=0xc000087fe0 sp=0xc000087fc8 pc=0x7ff792f82225 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000087fe8 sp=0xc000087fe0 pc=0x7ff792fdd9a1 created by runtime.gcenable in goroutine 1 runtime/mgc.go:205 +0xa5 goroutine 5 gp=0xc000003340 m=nil [finalizer wait]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000089e30 sp=0xc000089e10 pc=0x7ff792fd598e runtime.runfinq() runtime/mfinal.go:196 +0x107 fp=0xc000089fe0 sp=0xc000089e30 pc=0x7ff792f81207 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000089fe8 sp=0xc000089fe0 pc=0x7ff792fdd9a1 created by runtime.createfing in goroutine 1 runtime/mfinal.go:166 +0x3d goroutine 6 gp=0xc000003dc0 m=nil [chan receive]: runtime.gopark(0xc0000d5a40?, 0xc000510300?, 0x60?, 0x5f?, 0x7ff7930cc148?) runtime/proc.go:435 +0xce fp=0xc000075f18 sp=0xc000075ef8 pc=0x7ff792fd598e runtime.chanrecv(0xc00008c380, 0x0, 0x1) runtime/chan.go:664 +0x445 fp=0xc000075f90 sp=0xc000075f18 pc=0x7ff792f72d45 runtime.chanrecv1(0x7ff792fa4f40?, 0xc000075f76?) runtime/chan.go:506 +0x12 fp=0xc000075fb8 sp=0xc000075f90 pc=0x7ff792f728d2 runtime.unique_runtime_registerUniqueMapCleanup.func2(...) runtime/mgc.go:1796 runtime.unique_runtime_registerUniqueMapCleanup.gowrap1() runtime/mgc.go:1799 +0x2f fp=0xc000075fe0 sp=0xc000075fb8 pc=0x7ff792f854af runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000075fe8 sp=0xc000075fe0 pc=0x7ff792fdd9a1 created by unique.runtime_registerUniqueMapCleanup in goroutine 1 runtime/mgc.go:1794 +0x85 goroutine 7 gp=0xc0003da540 m=nil [GC worker (idle)]: runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000083f38 sp=0xc000083f18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008d960) runtime/mgc.go:1423 +0xe9 fp=0xc000083fc8 sp=0xc000083f38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc000083fe0 sp=0xc000083fc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000083fe8 sp=0xc000083fe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 8 gp=0xc0003da700 m=nil [GC worker (idle)]: runtime.gopark(0x4bde8a0df494?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000085f38 sp=0xc000085f18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008d960) runtime/mgc.go:1423 +0xe9 fp=0xc000085fc8 sp=0xc000085f38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc000085fe0 sp=0xc000085fc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000085fe8 sp=0xc000085fe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 9 gp=0xc0003da8c0 m=nil [GC worker (idle)]: runtime.gopark(0x4bde8a0df494?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc00047bf38 sp=0xc00047bf18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008d960) runtime/mgc.go:1423 +0xe9 fp=0xc00047bfc8 sp=0xc00047bf38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc00047bfe0 sp=0xc00047bfc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc00047bfe8 sp=0xc00047bfe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 18 gp=0xc0002081c0 m=nil [GC worker (idle)]: runtime.gopark(0x4bde8a0df494?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000477f38 sp=0xc000477f18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008d960) runtime/mgc.go:1423 +0xe9 fp=0xc000477fc8 sp=0xc000477f38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc000477fe0 sp=0xc000477fc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000477fe8 sp=0xc000477fe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 19 gp=0xc000208380 m=nil [GC worker (idle)]: runtime.gopark(0x4bde8a0df494?, 0x3?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000479f38 sp=0xc000479f18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008d960) runtime/mgc.go:1423 +0xe9 fp=0xc000479fc8 sp=0xc000479f38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc000479fe0 sp=0xc000479fc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000479fe8 sp=0xc000479fe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 20 gp=0xc000208540 m=nil [GC worker (idle)]: runtime.gopark(0x4bde8a0df494?, 0x0?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000213f38 sp=0xc000213f18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008d960) runtime/mgc.go:1423 +0xe9 fp=0xc000213fc8 sp=0xc000213f38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc000213fe0 sp=0xc000213fc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000213fe8 sp=0xc000213fe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 34 gp=0xc000484000 m=nil [GC worker (idle)]: runtime.gopark(0x4bde8a0df494?, 0x3?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc00020ff38 sp=0xc00020ff18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008d960) runtime/mgc.go:1423 +0xe9 fp=0xc00020ffc8 sp=0xc00020ff38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc00020ffe0 sp=0xc00020ffc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc00020ffe8 sp=0xc00020ffe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 21 gp=0xc000208700 m=nil [GC worker (idle)]: runtime.gopark(0x7ff7951fc600?, 0x1?, 0x0?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000215f38 sp=0xc000215f18 pc=0x7ff792fd598e runtime.gcBgMarkWorker(0xc00008d960) runtime/mgc.go:1423 +0xe9 fp=0xc000215fc8 sp=0xc000215f38 pc=0x7ff792f847a9 runtime.gcBgMarkStartWorkers.gowrap1() runtime/mgc.go:1339 +0x25 fp=0xc000215fe0 sp=0xc000215fc8 pc=0x7ff792f84685 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000215fe8 sp=0xc000215fe0 pc=0x7ff792fdd9a1 created by runtime.gcBgMarkStartWorkers in goroutine 1 runtime/mgc.go:1339 +0x105 goroutine 10 gp=0xc000484a80 m=nil [sync.WaitGroup.Wait]: runtime.gopark(0x0?, 0x0?, 0x60?, 0x0?, 0x0?) runtime/proc.go:435 +0xce fp=0xc000211e20 sp=0xc000211e00 pc=0x7ff792fd598e runtime.goparkunlock(...) runtime/proc.go:441 runtime.semacquire1(0xc00043fb00, 0x0, 0x1, 0x0, 0x18) runtime/sema.go:188 +0x22f fp=0xc000211e88 sp=0xc000211e20 pc=0x7ff792fb750f sync.runtime_SemacquireWaitGroup(0x0?) runtime/sema.go:110 +0x25 fp=0xc000211ec0 sp=0xc000211e88 pc=0x7ff792fd6f85 sync.(*WaitGroup).Wait(0x0?) sync/waitgroup.go:118 +0x48 fp=0xc000211ee8 sp=0xc000211ec0 pc=0x7ff792feb988 github.com/ollama/ollama/runner/llamarunner.(*Server).run(0xc00043fae0, {0x7ff7946e4a30, 0xc00033e2d0}) github.com/ollama/ollama/runner/llamarunner/runner.go:360 +0x4b fp=0xc000211fb8 sp=0xc000211ee8 pc=0x7ff79352922b github.com/ollama/ollama/runner/llamarunner.Execute.gowrap1() github.com/ollama/ollama/runner/llamarunner/runner.go:981 +0x28 fp=0xc000211fe0 sp=0xc000211fb8 pc=0x7ff79352e788 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc000211fe8 sp=0xc000211fe0 pc=0x7ff792fdd9a1 created by github.com/ollama/ollama/runner/llamarunner.Execute in goroutine 1 github.com/ollama/ollama/runner/llamarunner/runner.go:981 +0x4c5 goroutine 11 gp=0xc000484c40 m=nil [IO wait]: runtime.gopark(0x0?, 0xc00060d6a0?, 0x48?, 0xd7?, 0xc00060d74c?) runtime/proc.go:435 +0xce fp=0xc0004c58c8 sp=0xc0004c58a8 pc=0x7ff792fd598e runtime.netpollblock(0x38c?, 0x92f70406?, 0xf7?) runtime/netpoll.go:575 +0xf7 fp=0xc0004c5900 sp=0xc0004c58c8 pc=0x7ff792f9bdf7 internal/poll.runtime_pollWait(0x1eafa29c358, 0x72) runtime/netpoll.go:351 +0x85 fp=0xc0004c5920 sp=0xc0004c5900 pc=0x7ff792fd4b25 internal/poll.(*pollDesc).wait(0x7ff79319d5f7?, 0xc0004c5970?, 0x0) internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc0004c5948 sp=0xc0004c5920 pc=0x7ff79306bee7 internal/poll.execIO(0xc00060d6a0, 0x7ff794546098) internal/poll/fd_windows.go:177 +0x105 fp=0xc0004c59c0 sp=0xc0004c5948 pc=0x7ff79306d345 internal/poll.(*FD).Read(0xc00060d688, {0xc000091000, 0x1000, 0x1000}) internal/poll/fd_windows.go:438 +0x29b fp=0xc0004c5a60 sp=0xc0004c59c0 pc=0x7ff79306e01b net.(*netFD).Read(0xc00060d688, {0xc000091000?, 0xc0004c5ad0?, 0x7ff79306c3a5?}) net/fd_posix.go:55 +0x25 fp=0xc0004c5aa8 sp=0xc0004c5a60 pc=0x7ff7930e1325 net.(*conn).Read(0xc000078890, {0xc000091000?, 0x0?, 0x0?}) net/net.go:194 +0x45 fp=0xc0004c5af0 sp=0xc0004c5aa8 pc=0x7ff7930f0a45 net/http.(*connReader).Read(0xc000326090, {0xc000091000, 0x1000, 0x1000}) net/http/server.go:798 +0x159 fp=0xc0004c5b40 sp=0xc0004c5af0 pc=0x7ff7932de639 bufio.(*Reader).fill(0xc000193e60) bufio/bufio.go:113 +0x103 fp=0xc0004c5b78 sp=0xc0004c5b40 pc=0x7ff793107963 bufio.(*Reader).Peek(0xc000193e60, 0x4) bufio/bufio.go:152 +0x53 fp=0xc0004c5b98 sp=0xc0004c5b78 pc=0x7ff793107a93 net/http.(*conn).serve(0xc000632480, {0x7ff7946e49f8, 0xc0001c6090}) net/http/server.go:2137 +0x785 fp=0xc0004c5fb8 sp=0xc0004c5b98 pc=0x7ff7932e4425 net/http.(*Server).Serve.gowrap3() net/http/server.go:3454 +0x28 fp=0xc0004c5fe0 sp=0xc0004c5fb8 pc=0x7ff7932e9b88 runtime.goexit({}) runtime/asm_amd64.s:1700 +0x1 fp=0xc0004c5fe8 sp=0xc0004c5fe0 pc=0x7ff792fdd9a1 created by net/http.(*Server).Serve in goroutine 1 net/http/server.go:3454 +0x485 rax 0xffffffed2d005900 rbx 0x1ea9f1f9110 rcx 0x1eafd2dbf10 rdx 0x0 rdi 0x0 rsi 0x1ea9f293a80 rbp 0x41d58ff930 rsp 0x41d58ff680 r8 0x1eab4d10fc0 r9 0x1 r10 0x8000 r11 0x41d58ff690 r12 0x667 r13 0x1218 r14 0x1 r15 0x1eafd0e00b0 rip 0x7ffceb1fe53e rflags 0x10287 cs 0x33 fs 0x53 gs 0x2b time=2026-02-17T16:21:25.001+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server error" time=2026-02-17T16:21:26.003+01:00 level=INFO source=sched.go:493 msg="Load failed" model=F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e error="llama runner process has terminated: exit status 2" [GIN] 2026/02/17 - 16:21:26 | 500 | 24.9481468s | 127.0.0.1 | POST "/api/chat" [GIN] 2026/02/17 - 16:21:30 | 200 | 13.9638ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:21:37 | 200 | 127.8505ms | 127.0.0.1 | POST "/api/show" [GIN] 2026/02/17 - 16:21:40 | 200 | 10.4352ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:21:42 | 200 | 9.3882ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/17 - 16:21:42 | 200 | 126.1458ms | 127.0.0.1 | POST "/api/show" [GIN] 2026/02/17 - 16:21:42 | 200 | 118.4192ms | 127.0.0.1 | POST "/api/show" time=2026-02-17T16:21:42.956+01:00 level=INFO source=cpu_windows.go:148 msg=packages count=1 time=2026-02-17T16:21:42.957+01:00 level=INFO source=cpu_windows.go:195 msg="" package=0 cores=4 efficiency=0 threads=8 llama_model_loader: loaded meta data with 26 key-value pairs and 579 tensors from F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e (version GGUF V3 (latest)) llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output. llama_model_loader: - kv 0: general.architecture str = qwen2 llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.name str = DeepSeek R1 Distill Qwen 14B llama_model_loader: - kv 3: general.basename str = DeepSeek-R1-Distill-Qwen llama_model_loader: - kv 4: general.size_label str = 14B llama_model_loader: - kv 5: qwen2.block_count u32 = 48 llama_model_loader: - kv 6: qwen2.context_length u32 = 131072 llama_model_loader: - kv 7: qwen2.embedding_length u32 = 5120 llama_model_loader: - kv 8: qwen2.feed_forward_length u32 = 13824 llama_model_loader: - kv 9: qwen2.attention.head_count u32 = 40 llama_model_loader: - kv 10: qwen2.attention.head_count_kv u32 = 8 llama_model_loader: - kv 11: qwen2.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 12: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 13: general.file_type u32 = 15 llama_model_loader: - kv 14: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 15: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 16: tokenizer.ggml.tokens arr[str,152064] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 18: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 151646 llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151643 llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 24: tokenizer.chat_template str = {% if not add_generation_prompt is de... llama_model_loader: - kv 25: general.quantization_version u32 = 2 llama_model_loader: - type f32: 241 tensors llama_model_loader: - type q4_K: 289 tensors llama_model_loader: - type q6_K: 49 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 8.37 GiB (4.87 BPW) load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect load: printing all EOG tokens: load: - 151643 ('<｜end▁of▁sentence｜>') load: - 151662 ('<|fim_pad|>') load: - 151663 ('<|repo_name|>') load: - 151664 ('<|file_sep|>') load: special tokens cache size = 22 load: token to piece cache size = 0.9310 MB print_info: arch = qwen2 print_info: vocab_only = 1 print_info: no_alloc = 0 print_info: model type = ?B print_info: model params = 14.77 B print_info: general.name = DeepSeek R1 Distill Qwen 14B print_info: vocab type = BPE print_info: n_vocab = 152064 print_info: n_merges = 151387 print_info: BOS token = 151646 '<｜begin▁of▁sentence｜>' print_info: EOS token = 151643 '<｜end▁of▁sentence｜>' print_info: EOT token = 151643 '<｜end▁of▁sentence｜>' print_info: PAD token = 151643 '<｜end▁of▁sentence｜>' print_info: LF token = 198 'Ċ' print_info: FIM PRE token = 151659 '<|fim_prefix|>' print_info: FIM SUF token = 151661 '<|fim_suffix|>' print_info: FIM MID token = 151660 '<|fim_middle|>' print_info: FIM PAD token = 151662 '<|fim_pad|>' print_info: FIM REP token = 151663 '<|repo_name|>' print_info: FIM SEP token = 151664 '<|file_sep|>' print_info: EOG token = 151643 '<｜end▁of▁sentence｜>' print_info: EOG token = 151662 '<|fim_pad|>' print_info: EOG token = 151663 '<|repo_name|>' print_info: EOG token = 151664 '<|file_sep|>' print_info: max token length = 256 llama_model_load: vocab only - skipping tensors time=2026-02-17T16:21:43.281+01:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\robob\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --model F:\\OllamaModels\\blobs\\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e --port 34156" time=2026-02-17T16:21:43.282+01:00 level=INFO source=sched.go:466 msg="system memory" total="15.9 GiB" free="9.7 GiB" free_swap="17.0 GiB" time=2026-02-17T16:21:43.282+01:00 level=INFO source=server.go:498 msg="loading model" "model layers"=49 requested=0 time=2026-02-17T16:21:43.283+01:00 level=INFO source=device.go:245 msg="model weights" device=CPU size="8.0 GiB" time=2026-02-17T16:21:43.283+01:00 level=INFO source=device.go:256 msg="kv cache" device=CPU size="768.0 MiB" time=2026-02-17T16:21:43.283+01:00 level=INFO source=device.go:272 msg="total memory" size="8.7 GiB" time=2026-02-17T16:21:43.374+01:00 level=INFO source=runner.go:965 msg="starting go runner" load_backend: loaded CPU backend from C:\Users\robob\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll time=2026-02-17T16:21:43.443+01:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 compiler=cgo(clang) time=2026-02-17T16:21:43.444+01:00 level=INFO source=runner.go:1001 msg="Server listening on 127.0.0.1:34156" time=2026-02-17T16:21:43.445+01:00 level=INFO source=runner.go:895 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Auto KvSize:4096 KvCacheType: NumThreads:4 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2026-02-17T16:21:43.446+01:00 level=INFO source=server.go:1350 msg="waiting for llama runner to start responding" time=2026-02-17T16:21:43.446+01:00 level=INFO source=server.go:1384 msg="waiting for server to become available" status="llm server loading model" llama_model_loader: loaded meta data with 26 key-value pairs and 579 tensors from F:\OllamaModels\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e (version GGUF V3 (latest)) llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output. llama_model_loader: - kv 0: general.architecture str = qwen2 llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.name str = DeepSeek R1 Distill Qwen 14B llama_model_loader: - kv 3: general.basename str = DeepSeek-R1-Distill-Qwen llama_model_loader: - kv 4: general.size_label str = 14B llama_model_loader: - kv 5: qwen2.block_count u32 = 48 llama_model_loader: - kv 6: qwen2.context_length u32 = 131072 llama_model_loader: - kv 7: qwen2.embedding_length u32 = 5120 llama_model_loader: - kv 8: qwen2.feed_forward_length u32 = 13824 llama_model_loader: - kv 9: qwen2.attention.head_count u32 = 40 llama_model_loader: - kv 10: qwen2.attention.head_count_kv u32 = 8 llama_model_loader: - kv 11: qwen2.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 12: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 13: general.file_type u32 = 15 llama_model_loader: - kv 14: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 15: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 16: tokenizer.ggml.tokens arr[str,152064] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 18: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 151646 llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151643 llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 24: tokenizer.chat_template str = {% if not add_generation_prompt is de... llama_model_loader: - kv 25: general.quantization_version u32 = 2 llama_model_loader: - type f32: 241 tensors llama_model_loader: - type q4_K: 289 tensors llama_model_loader: - type q6_K: 49 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 8.37 GiB (4.87 BPW) load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect load: printing all EOG tokens: load: - 151643 ('<｜end▁of▁sentence｜>') load: - 151662 ('<|fim_pad|>') load: - 151663 ('<|repo_name|>') load: - 151664 ('<|file_sep|>') load: special tokens cache size = 22 load: token to piece cache size = 0.9310 MB print_info: arch = qwen2 print_info: vocab_only = 0 print_info: no_alloc = 0 print_info: n_ctx_train = 131072 print_info: n_embd = 5120 print_info: n_embd_inp = 5120 print_info: n_layer = 48 print_info: n_head = 40 print_info: n_head_kv = 8 print_info: n_rot = 128 print_info: n_swa = 0 print_info: is_swa_any = 0 print_info: n_embd_head_k = 128 print_info: n_embd_head_v = 128 print_info: n_gqa = 5 print_info: n_embd_k_gqa = 1024 print_info: n_embd_v_gqa = 1024 print_info: f_norm_eps = 0.0e+00 print_info: f_norm_rms_eps = 1.0e-05 print_info: f_clamp_kqv = 0.0e+00 print_info: f_max_alibi_bias = 0.0e+00 print_info: f_logit_scale = 0.0e+00 print_info: f_attn_scale = 0.0e+00 print_info: n_ff = 13824 print_info: n_expert = 0 print_info: n_expert_used = 0 print_info: n_expert_groups = 0 print_info: n_group_used = 0 print_info: causal attn = 1 print_info: pooling type = -1 print_info: rope type = 2 print_info: rope scaling = linear print_info: freq_base_train = 1000000.0 print_info: freq_scale_train = 1 print_info: n_ctx_orig_yarn = 131072 print_info: rope_yarn_log_mul= 0.0000 print_info: rope_finetuned = unknown print_info: model type = 14B print_info: model params = 14.77 B print_info: general.name = DeepSeek R1 Distill Qwen 14B print_info: vocab type = BPE print_info: n_vocab = 152064 print_info: n_merges = 151387 print_info: BOS token = 151646 '<｜begin▁of▁sentence｜>' print_info: EOS token = 151643 '<｜end▁of▁sentence｜>' print_info: EOT token = 151643 '<｜end▁of▁sentence｜>' print_info: PAD token = 151643 '<｜end▁of▁sentence｜>' print_info: LF token = 198 'Ċ' print_info: FIM PRE token = 151659 '<|fim_prefix|>' print_info: FIM SUF token = 151661 '<|fim_suffix|>' print_info: FIM MID token = 151660 '<|fim_middle|>' print_info: FIM PAD token = 151662 '<|fim_pad|>' print_info: FIM REP token = 151663 '<|repo_name|>' print_info: FIM SEP token = 151664 '<|file_sep|>' print_info: EOG token = 151643 '<｜end▁of▁sentence｜>' print_info: EOG token = 151662 '<|fim_pad|>' print_info: EOG token = 151663 '<|repo_name|>' print_info: EOG token = 151664 '<|file_sep|>' print_info: max token length = 256 load_tensors: loading model tensors, this can take a while... (mmap = false) load_tensors: CPU model buffer size = 8566.04 MiB llama_context: constructing llama_context llama_context: n_seq_max = 1 llama_context: n_ctx = 4096 llama_context: n_ctx_seq = 4096 llama_context: n_batch = 512 llama_context: n_ubatch = 512 llama_context: causal_attn = 1 llama_context: flash_attn = auto llama_context: kv_unified = false llama_context: freq_base = 1000000.0 llama_context: freq_scale = 1 llama_context: n_ctx_seq (4096) < n_ctx_train (131072) -- the full capacity of the model will not be utilized llama_context: CPU output buffer size = 0.60 MiB llama_kv_cache: CPU KV buffer size = 768.00 MiB llama_kv_cache: size = 768.00 MiB ( 4096 cells, 48 layers, 1/1 seqs), K (f16): 384.00 MiB, V (f16): 384.00 MiB llama_context: Flash Attention was auto, set to enabled llama_context: CPU compute buffer size = 349.01 MiB llama_context: graph nodes = 1639 llama_context: graph splits = 1 time=2026-02-17T16:22:05.502+01:00 level=INFO source=server.go:1388 msg="llama runner started in 22.22 seconds" time=2026-02-17T16:22:05.502+01:00 level=INFO source=sched.go:540 msg="loaded runners" count=1 time=2026-02-17T16:22:05.502+01:00 level=INFO source=server.go:1350 msg="waiting for llama runner to start responding" time=2026-02-17T16:22:05.503+01:00 level=INFO source=server.go:1388 msg="llama runner started in 22.22 seconds"

GiteaMirror referenced this issue

2026-04-22 12:35:59 -05:00

[GH-ISSUE #9301] Add response to logs #31828

GiteaMirror referenced this issue

2026-04-28 23:46:19 -05:00

[GH-ISSUE #9301] Add response to logs #52579

GiteaMirror referenced this issue

2026-05-04 12:35:44 -05:00

[GH-ISSUE #9301] Add response to logs #68124

GiteaMirror referenced this issue

2026-05-09 19:02:06 -05:00

[GH-ISSUE #9301] Add response to logs #83753

Sign in to join this conversation.

Branches Tags

main

dhiltgen/llama-runner

parth-migrate-pi

codex/make-integration-hidden-and-lunchable

hoyyeva/migrate-pi

hoyyeva/opencode-thinking

hoyyeva/anthropic-local-image-path

parth-launch-codex-app

hoyyeva/anthropic-reference-images-path

parth-anthropic-reference-images-path

brucemacd/download-before-remove

hoyyeva/editor-config-repair

parth-mlx-decode-checkpoints

hoyyeva/qwen

parth/hide-claude-desktop-till-release

parth-add-claude-code-autoinstall

release_v0.22.0

pdevine/manifest-list

codex/fix-codex-model-metadata-warning

pdevine/addressable-manifest

brucemacd/launch-fetch-reccomended

jmorganca/llama-compat

launch-copilot-cli

release_v0.20.7

parth-auto-save-backup

parth-test

jmorganca/gemma4-audio-replacements

fix-manifest-digest-on-pull

hoyyeva/vscode-improve

brucemacd/install-server-wait

parth/update-claude-docs

brucemac/start-ap-install

pdevine/mlx-update

pdevine/qwen35_vision

drifkin/api-show-fallback

mintlify/image-generation-1773352582

hoyyeva/server-context-length-local-config

jmorganca/faster-reptition-penalties

jmorganca/convert-nemotron

parth-pi-thinking

pdevine/sampling-penalties

jmorganca/fix-create-quantization-memory

dongchen/resumable_transfer_fix

pdevine/sampling-cache-error

jessegross/mlx-usage

hoyyeva/openclaw-config

hoyyeva/app-html

pdevine/qwen3next

brucemacd/sign-sh-install

brucemacd/tui-update

brucemacd/usage-api

jmorganca/launch-empty

fix-app-dist-embed

mxyng/mlx-compile

mxyng/mlx-quant

mxyng/mlx-glm4.7

mxyng/mlx

brucemacd/simplify-model-picker

jmorganca/qwen3-concurrent

fix-glm-4.7-flash-mla-config

drifkin/qwen3-coder-opening-tag

brucemacd/usage-cli

fix-cuda12-fattn-shmem

ollama-imagegen-docs

parth/fix-multiline-inputs

brucemacd/config-docs

mxyng/model-files

mxyng/simple-execute

fix-imagegen-ollama-models

mxyng/async-upload

jmorganca/lazy-no-dtype-changes

imagegen-auto-detect-create

parth/decrease-concurrent-download-hf

fix-mlx-quantize-init

jmorganca/x-cleanup

usage

imagegen-readme

jmorganca/glm-image

mlx-gpu-cd

jmorganca/imagegen-modelfile

parth/agent-skills

parth/agent-allowlist

parth/signed-in-offline

parth/agents

parth/fix-context-chopping

improve-cloud-flow

parth/add-models-websearch

parth/prompt-renderer-mcp

jmorganca/native-settings

jmorganca/download-stream-hash

jmorganca/client2-rebased

brucemacd/oai-chat-req-multipart

jessegross/multi_chunk_reserve

grace/additional-omit-empty

grace/mistral-3-large

mxyng/tokenizer2

mxyng/tokenizer

jessegross/flash

hoyyeva/windows-nacked-app

mxyng/cleanup-attention

grace/deepseek-parser

hoyyeva/remember-unsent-prompt

parth/add-lfs-pointer-error-conversion

parth/olmo2-test2

hoyyeva/ollama-launchagent-plist

nicole/olmo-model

parth/olmo-test

mxyng/remove-embedded

parth/render-template

jmorganca/intellect-3

parth/remove-prealloc-linter

jmorganca/cmd-eval

nicole/nomic-embed-text-fix

mxyng/lint-2

hoyyeva/add-gemini-3-pro-preview

hoyyeva/load-model-list

mxyng/expand-path

mxyng/environ-2

hoyyeva/deeplink-json-encoding

parth/improve-tool-calling-tests

hoyyeva/conversation

hoyyeva/assistant-edit-response

hoyyeva/thinking

origin/brucemacd/invalid-char-i-err

parth/improve-tool-calling

jmorganca/required-omitempty

grace/qwen3-vl-tests

mxyng/iter-client

parth/docs-readme

nicole/embed-test

pdevine/integration-benchstat

parth/remove-generate-cmd

parth/add-toolcall-id

mxyng/server-tests

jmorganca/glm-4.6

jmorganca/gin-h-compat

drifkin/stable-tool-args

pdevine/qwen3-more-thinking

parth/add-websearch-client

nicole/websearch_local

jmorganca/qwen3-coder-updates

grace/deepseek-v3-migration-tests

mxyng/fix-create

jmorganca/cloud-errors

pdevine/parser-tidy

revert-12233-parth/simplify-entrypoints-runner

parth/enable-so-gpt-oss

brucemacd/qwen3vl

jmorganca/readme-simplify

parth/gpt-oss-structured-outputs

revert-12039-jmorganca/tools-braces

mxyng/embeddings

mxyng/gguf

mxyng/benchmark

mxyng/types-null

parth/move-parsing

mxyng/gemma2

jmorganca/docs

mxyng/16-bit

mxyng/create-stdin

pdevine/authorizedkeys

mxyng/quant

parth/opt-in-error-context-window

brucemacd/cache-models

brucemacd/runner-completion

jmorganca/llama-update-6

brucemacd/benchmark-list

brucemacd/partial-read-caps

parth/deepseek-r1-tools

mxyng/omit-array

parth/tool-prefix-temp

brucemacd/runner-test

jmorganca/qwen25vl

brucemacd/model-forward-test-ext

parth/python-function-parsing

jmorganca/cuda-compression-none

drifkin/num-parallel

drifkin/chat-truncation-fix

jmorganca/sync

parth/python-tools-calling

drifkin/array-head-count

brucemacd/create-no-loop

parth/server-enable-content-stream-with-tools

qwen25omni

mxyng/v3

brucemacd/ropeconfig

jmorganca/silence-tokenizer

parth/sample-so-test

parth/sampling-structured-outputs

brucemacd/doc-go-engine

parth/constrained-sampling-json

jmorganca/mistral-wip

brucemacd/mistral-small-convert

parth/sample-unmarshal-json-for-params

brucemacd/jomorganca/mistral

pdevine/bfloat16

jmorganca/mistral

brucemacd/mistral

pdevine/logging

parth/sample-correctness-fix

parth/sample-fix-sorting

jmorgan/sample-fix-sorting-extras

jmorganca/temp-0-images

brucemacd/parallel-embed-models

brucemacd/shim-grammar

jmorganca/fix-gguf-error

bmizerany/nameswork

jmorganca/faster-releases

bmizerany/validatenames

brucemacd/err-no-vocab

brucemacd/rope-config

brucemacd/err-hint

brucemacd/qwen2_5

brucemacd/logprobs

brucemacd/new_runner_graph_bench

progress-flicker

brucemacd/forward-test

brucemacd/go_qwen2

pdevine/gemma2

jmorganca/add-missing-symlink-eval

mxyng/next-debug

parth/set-context-size-openai

brucemacd/next-bpe-bench

brucemacd/next-bpe-test

brucemacd/new_runner_e2e

brucemacd/new_runner_qwen2

pdevine/convert-cohere2

brucemacd/convert-cli

parth/log-probs

mxyng/next-mlx

mxyng/cmd-history

parth/templating

parth/tokenize-detokenize

brucemacd/check-key-register

bmizerany/grammar

jmorganca/vendor-081b29bd

mxyng/func-checks

jmorganca/fix-null-format

parth/fix-default-to-warn-json

jmorganca/qwen2vl

jmorganca/no-concat

parth/cmd-cleanup-SO

brucemacd/check-key-register-structured-err

parth/openai-stream-usage

parth/fix-referencing-so

stream-tools-stop

jmorganca/degin-1

brucemacd/install-path-clean

brucemacd/push-name-validation

brucemacd/browser-key-register

jmorganca/openai-fix-first-message

jmorganca/fix-proxy

jessegross/sample

parth/disallow-streaming-tools

dhiltgen/remove_submodule

jmorganca/ga

jmorganca/mllama

pdevine/newlines

pdevine/geems-2b

jmorganca/llama-bump

mxyng/modelname-7

mxyng/gin-slog

mxyng/modelname-6

jyan/convert-prog

jyan/quant5

paligemma-support

pdevine/import-docs

jmorganca/openai-context

jyan/paligemma

jyan/p2

jyan/palitest

bmizerany/embedspeedup

jmorganca/llama-vit

brucemacd/allow-ollama

royh/ep-methods

royh/whisper

mxyng/api-models

mxyng/fix-memory

jyan/q4_4/8

jyan/ollama-v

royh/stream-tools

roy-embed-parallel

bmizerany/hrm

revert-5963-revert-5924-mxyng/llama3.1-rope

royh/embed-viz

jyan/local2

jyan/auth

jyan/local

jyan/parse-temp

jmorganca/template-mistral

jyan/reord-g

royh-openai-suffixdocs

royh-imgembed

royh-embed-parallel

jyan/quant4

royh-precision

jyan/progress

pdevine/fix-template

jyan/quant3

pdevine/ggla

mxyng/update-registry-domain

jmorganca/ggml-static

mxyng/create-context

jyan/v0.146

mxyng/layers-from-files

build_dist

bmizerany/noseek

royh-ls

royh-name

timeout

mxyng/server-timestamp

bmizerany/nosillyggufslurps

royh-params

jmorganca/llama-cpp-7c26775

royh-openai-delete

royh-show-rigid

jmorganca/enable-fa

jmorganca/no-error-template

jyan/format

royh-testdelete

bmizerany/fastverify

language_support

pdevine/ps-glitches

brucemacd/tokenize

bruce/iq-quants

bmizerany/filepathwithcoloninhost

mxyng/split-bin

bmizerany/client-registry

jmorganca/if-none-match

native

jmorganca/native

jmorganca/batch-embeddings

jmorganca/initcmake

jmorganca/mm

pdevine/showggmlinfo

modenameenforcealphanum

bmizerany/modenameenforcealphanum

jmorganca/done-reason

jmorganca/llama-cpp-8960fe8

ollama.com

bmizerany/filepathnobuild

bmizerany/types/model/defaultfix

rmdisplaylong

nogogen

bmizerany/x

modelfile-readme

bmizerany/replacecolon

jmorganca/limit

jmorganca/execstack

jmorganca/replace-assets

mxyng/tune-concurrency

jmorganca/testing

whitespace-detection

jmorganca/options

upgrade-all

scratch

cuda-search

mattw/airenamer

mattw/allmodelsonhuggingface

mattw/quantcontext

mattw/whatneedstorun

brucemacd/llama-mem-calc

mattw/faq-context

mattw/communitylinks

mattw/noprune

mattw/python-functioncalling

rename

mxyng/install

pulse

remove-first

editor

mattw/selfqueryingretrieval

cgo

mattw/howtoquant

api

matt/streamingapi

format-config

mxyng/extra-args

shell

update-nous-hermes

cp-model

upload-progress

fix-unknown-model

fix-model-names

delete-fix

insecure-registry

ls

deletemodels

progressbar

readme-updates

license-layers

skip-list

list-models

modelpath

matt/examplemodelfiles

distribution

go-opts

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: github-starred/ollama#9301