github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #9245] did not use gpu #31784

New Issue
Closed
opened 2026-04-22 12:33:10 -05:00 by GiteaMirror · 8 comments
GiteaMirror commented 2026-04-22 12:33:10 -05:00
Owner
Copy Link

Originally created by @Hsq12138 on GitHub (Feb 20, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/9245

What is the issue?

I found that when running the deepseek14b model, it only uses the CPU, with the GPU usage at 0 in the task manager.

Relevant log output

2025/02/20 15:45:52 routes.go:1186: INFO server config env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_DEBUG:false OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://0.0.0.0:11434 OLLAMA_INTEL_GPU:false 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:D:\\ollama\\models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:0 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://*] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES:]"
time=2025-02-20T15:45:52.074+08:00 level=INFO source=images.go:432 msg="total blobs: 7"
time=2025-02-20T15:45:52.075+08:00 level=INFO source=images.go:439 msg="total unused blobs removed: 0"
time=2025-02-20T15:45:52.075+08:00 level=INFO source=routes.go:1237 msg="Listening on [::]:11434 (version 0.5.11)"
time=2025-02-20T15:45:52.075+08:00 level=INFO source=gpu.go:217 msg="looking for compatible GPUs"
time=2025-02-20T15:45:52.075+08:00 level=INFO source=gpu_windows.go:167 msg=packages count=1
time=2025-02-20T15:45:52.075+08:00 level=INFO source=gpu_windows.go:214 msg="" package=0 cores=6 efficiency=0 threads=12
time=2025-02-20T15:45:52.167+08:00 level=INFO source=gpu.go:319 msg="detected OS VRAM overhead" id=GPU-ecc0382b-7d7c-7b61-8572-a21b10ac9fcd library=cuda compute=8.9 driver=12.8 name="NVIDIA GeForce RTX 4080 SUPER" overhead="176.2 MiB"
time=2025-02-20T15:45:52.168+08:00 level=INFO source=types.go:130 msg="inference compute" id=GPU-ecc0382b-7d7c-7b61-8572-a21b10ac9fcd library=cuda variant=v12 compute=8.9 driver=12.8 name="NVIDIA GeForce RTX 4080 SUPER" total="16.0 GiB" available="14.7 GiB"
[GIN] 2025/02/20 - 15:49:16 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2025/02/20 - 15:49:16 | 200 |       525.4µs |       127.0.0.1 | GET      "/api/tags"
[GIN] 2025/02/20 - 15:49:31 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2025/02/20 - 15:49:31 | 200 |     15.0919ms |       127.0.0.1 | POST     "/api/show"
time=2025-02-20T15:49:31.524+08:00 level=INFO source=sched.go:714 msg="new model will fit in available VRAM in single GPU, loading" model=D:\ollama\models\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e gpu=GPU-ecc0382b-7d7c-7b61-8572-a21b10ac9fcd parallel=4 available=15652315136 required="10.8 GiB"
time=2025-02-20T15:49:31.536+08:00 level=INFO source=server.go:100 msg="system memory" total="95.8 GiB" free="86.6 GiB" free_swap="88.1 GiB"
time=2025-02-20T15:49:31.537+08:00 level=INFO source=memory.go:356 msg="offload to cuda" layers.requested=-1 layers.model=49 layers.offload=49 layers.split="" memory.available="[14.6 GiB]" memory.gpu_overhead="0 B" memory.required.full="10.8 GiB" memory.required.partial="10.8 GiB" memory.required.kv="1.5 GiB" memory.required.allocations="[10.8 GiB]" memory.weights.total="8.9 GiB" memory.weights.repeating="8.3 GiB" memory.weights.nonrepeating="609.1 MiB" memory.graph.full="676.0 MiB" memory.graph.partial="916.1 MiB"
time=2025-02-20T15:49:31.543+08:00 level=INFO source=server.go:380 msg="starting llama server" cmd="C:\\Users\\zrway\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --model D:\\ollama\\models\\blobs\\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e --ctx-size 8192 --batch-size 512 --n-gpu-layers 49 --threads 6 --no-mmap --parallel 4 --port 50269"
time=2025-02-20T15:49:31.594+08:00 level=INFO source=sched.go:449 msg="loaded runners" count=1
time=2025-02-20T15:49:31.594+08:00 level=INFO source=server.go:557 msg="waiting for llama runner to start responding"
time=2025-02-20T15:49:31.594+08:00 level=INFO source=server.go:591 msg="waiting for server to become available" status="llm server error"
time=2025-02-20T15:49:31.610+08:00 level=INFO source=runner.go:936 msg="starting go runner"
time=2025-02-20T15:49:31.616+08:00 level=INFO source=runner.go:937 msg=system info="CPU : LLAMAFILE = 1 | CPU : LLAMAFILE = 1 | cgo(clang)" threads=6
time=2025-02-20T15:49:31.616+08:00 level=INFO source=runner.go:995 msg="Server listening on 127.0.0.1:50269"
ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-alderlake.dll
ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll
ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-icelake.dll
ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-sandybridge.dll
ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-skylakex.dll
llama_model_loader: loaded meta data with 26 key-value pairs and 579 tensors from D:\ollama\models\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
llm_load_vocab: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
llm_load_vocab: special tokens cache size = 22
llm_load_vocab: token to piece cache size = 0.9310 MB
llm_load_print_meta: format           = GGUF V3 (latest)
llm_load_print_meta: arch             = qwen2
llm_load_print_meta: vocab type       = BPE
llm_load_print_meta: n_vocab          = 152064
llm_load_print_meta: n_merges         = 151387
llm_load_print_meta: vocab_only       = 0
llm_load_print_meta: n_ctx_train      = 131072
llm_load_print_meta: n_embd           = 5120
llm_load_print_meta: n_layer          = 48
llm_load_print_meta: n_head           = 40
llm_load_print_meta: n_head_kv        = 8
llm_load_print_meta: n_rot            = 128
llm_load_print_meta: n_swa            = 0
llm_load_print_meta: n_embd_head_k    = 128
llm_load_print_meta: n_embd_head_v    = 128
llm_load_print_meta: n_gqa            = 5
llm_load_print_meta: n_embd_k_gqa     = 1024
llm_load_print_meta: n_embd_v_gqa     = 1024
llm_load_print_meta: f_norm_eps       = 0.0e+00
llm_load_print_meta: f_norm_rms_eps   = 1.0e-05
llm_load_print_meta: f_clamp_kqv      = 0.0e+00
llm_load_print_meta: f_max_alibi_bias = 0.0e+00
llm_load_print_meta: f_logit_scale    = 0.0e+00
llm_load_print_meta: n_ff             = 13824
llm_load_print_meta: n_expert         = 0
llm_load_print_meta: n_expert_used    = 0
llm_load_print_meta: causal attn      = 1
llm_load_print_meta: pooling type     = 0
llm_load_print_meta: rope type        = 2
llm_load_print_meta: rope scaling     = linear
llm_load_print_meta: freq_base_train  = 1000000.0
llm_load_print_meta: freq_scale_train = 1
llm_load_print_meta: n_ctx_orig_yarn  = 131072
llm_load_print_meta: rope_finetuned   = unknown
llm_load_print_meta: ssm_d_conv       = 0
llm_load_print_meta: ssm_d_inner      = 0
llm_load_print_meta: ssm_d_state      = 0
llm_load_print_meta: ssm_dt_rank      = 0
llm_load_print_meta: ssm_dt_b_c_rms   = 0
llm_load_print_meta: model type       = 14B
llm_load_print_meta: model ftype      = Q4_K - Medium
llm_load_print_meta: model params     = 14.77 B
llm_load_print_meta: model size       = 8.37 GiB (4.87 BPW) 
llm_load_print_meta: general.name     = DeepSeek R1 Distill Qwen 14B
llm_load_print_meta: BOS token        = 151646 '<｜begin▁of▁sentence｜>'
llm_load_print_meta: EOS token        = 151643 '<｜end▁of▁sentence｜>'
llm_load_print_meta: EOT token        = 151643 '<｜end▁of▁sentence｜>'
llm_load_print_meta: PAD token        = 151643 '<｜end▁of▁sentence｜>'
llm_load_print_meta: LF token         = 148848 'ÄĬ'
llm_load_print_meta: FIM PRE token    = 151659 '<|fim_prefix|>'
llm_load_print_meta: FIM SUF token    = 151661 '<|fim_suffix|>'
llm_load_print_meta: FIM MID token    = 151660 '<|fim_middle|>'
llm_load_print_meta: FIM PAD token    = 151662 '<|fim_pad|>'
llm_load_print_meta: FIM REP token    = 151663 '<|repo_name|>'
llm_load_print_meta: FIM SEP token    = 151664 '<|file_sep|>'
llm_load_print_meta: EOG token        = 151643 '<｜end▁of▁sentence｜>'
llm_load_print_meta: EOG token        = 151662 '<|fim_pad|>'
llm_load_print_meta: EOG token        = 151663 '<|repo_name|>'
llm_load_print_meta: EOG token        = 151664 '<|file_sep|>'
llm_load_print_meta: max token length = 256
llm_load_tensors:          CPU model buffer size =  8566.04 MiB
time=2025-02-20T15:49:31.845+08:00 level=INFO source=server.go:591 msg="waiting for server to become available" status="llm server loading model"
llama_new_context_with_model: n_seq_max     = 4
llama_new_context_with_model: n_ctx         = 8192
llama_new_context_with_model: n_ctx_per_seq = 2048
llama_new_context_with_model: n_batch       = 2048
llama_new_context_with_model: n_ubatch      = 512
llama_new_context_with_model: flash_attn    = 0
llama_new_context_with_model: freq_base     = 1000000.0
llama_new_context_with_model: freq_scale    = 1
llama_new_context_with_model: n_ctx_per_seq (2048) < n_ctx_train (131072) -- the full capacity of the model will not be utilized
llama_kv_cache_init: kv_size = 8192, offload = 1, type_k = 'f16', type_v = 'f16', n_layer = 48, can_shift = 1
llama_kv_cache_init:        CPU KV buffer size =  1536.00 MiB
llama_new_context_with_model: KV self size  = 1536.00 MiB, K (f16):  768.00 MiB, V (f16):  768.00 MiB
llama_new_context_with_model:        CPU  output buffer size =     2.40 MiB
llama_new_context_with_model:        CPU compute buffer size =   696.01 MiB
llama_new_context_with_model: graph nodes  = 1686
llama_new_context_with_model: graph splits = 1
[GIN] 2025/02/20 - 15:49:38 | 200 |    7.1239561s |       127.0.0.1 | POST     "/api/generate"
time=2025-02-20T15:49:38.605+08:00 level=INFO source=server.go:596 msg="llama runner started in 7.01 seconds"
[GIN] 2025/02/20 - 15:49:54 | 200 |     7.687977s |       127.0.0.1 | POST     "/api/chat"

OS

Windows

GPU

Nvidia

CPU

Intel

Ollama version

0.5.11

Originally created by @Hsq12138 on GitHub (Feb 20, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/9245 ### What is the issue? I found that when running the deepseek14b model, it only uses the CPU, with the GPU usage at 0 in the task manager. ### Relevant log output ```shell 2025/02/20 15:45:52 routes.go:1186: INFO server config env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_DEBUG:false OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://0.0.0.0:11434 OLLAMA_INTEL_GPU:false 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:D:\\ollama\\models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:0 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://*] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES:]" time=2025-02-20T15:45:52.074+08:00 level=INFO source=images.go:432 msg="total blobs: 7" time=2025-02-20T15:45:52.075+08:00 level=INFO source=images.go:439 msg="total unused blobs removed: 0" time=2025-02-20T15:45:52.075+08:00 level=INFO source=routes.go:1237 msg="Listening on [::]:11434 (version 0.5.11)" time=2025-02-20T15:45:52.075+08:00 level=INFO source=gpu.go:217 msg="looking for compatible GPUs" time=2025-02-20T15:45:52.075+08:00 level=INFO source=gpu_windows.go:167 msg=packages count=1 time=2025-02-20T15:45:52.075+08:00 level=INFO source=gpu_windows.go:214 msg="" package=0 cores=6 efficiency=0 threads=12 time=2025-02-20T15:45:52.167+08:00 level=INFO source=gpu.go:319 msg="detected OS VRAM overhead" id=GPU-ecc0382b-7d7c-7b61-8572-a21b10ac9fcd library=cuda compute=8.9 driver=12.8 name="NVIDIA GeForce RTX 4080 SUPER" overhead="176.2 MiB" time=2025-02-20T15:45:52.168+08:00 level=INFO source=types.go:130 msg="inference compute" id=GPU-ecc0382b-7d7c-7b61-8572-a21b10ac9fcd library=cuda variant=v12 compute=8.9 driver=12.8 name="NVIDIA GeForce RTX 4080 SUPER" total="16.0 GiB" available="14.7 GiB" [GIN] 2025/02/20 - 15:49:16 | 200 | 0s | 127.0.0.1 | HEAD "/" [GIN] 2025/02/20 - 15:49:16 | 200 | 525.4µs | 127.0.0.1 | GET "/api/tags" [GIN] 2025/02/20 - 15:49:31 | 200 | 0s | 127.0.0.1 | HEAD "/" [GIN] 2025/02/20 - 15:49:31 | 200 | 15.0919ms | 127.0.0.1 | POST "/api/show" time=2025-02-20T15:49:31.524+08:00 level=INFO source=sched.go:714 msg="new model will fit in available VRAM in single GPU, loading" model=D:\ollama\models\blobs\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e gpu=GPU-ecc0382b-7d7c-7b61-8572-a21b10ac9fcd parallel=4 available=15652315136 required="10.8 GiB" time=2025-02-20T15:49:31.536+08:00 level=INFO source=server.go:100 msg="system memory" total="95.8 GiB" free="86.6 GiB" free_swap="88.1 GiB" time=2025-02-20T15:49:31.537+08:00 level=INFO source=memory.go:356 msg="offload to cuda" layers.requested=-1 layers.model=49 layers.offload=49 layers.split="" memory.available="[14.6 GiB]" memory.gpu_overhead="0 B" memory.required.full="10.8 GiB" memory.required.partial="10.8 GiB" memory.required.kv="1.5 GiB" memory.required.allocations="[10.8 GiB]" memory.weights.total="8.9 GiB" memory.weights.repeating="8.3 GiB" memory.weights.nonrepeating="609.1 MiB" memory.graph.full="676.0 MiB" memory.graph.partial="916.1 MiB" time=2025-02-20T15:49:31.543+08:00 level=INFO source=server.go:380 msg="starting llama server" cmd="C:\\Users\\zrway\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --model D:\\ollama\\models\\blobs\\sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e --ctx-size 8192 --batch-size 512 --n-gpu-layers 49 --threads 6 --no-mmap --parallel 4 --port 50269" time=2025-02-20T15:49:31.594+08:00 level=INFO source=sched.go:449 msg="loaded runners" count=1 time=2025-02-20T15:49:31.594+08:00 level=INFO source=server.go:557 msg="waiting for llama runner to start responding" time=2025-02-20T15:49:31.594+08:00 level=INFO source=server.go:591 msg="waiting for server to become available" status="llm server error" time=2025-02-20T15:49:31.610+08:00 level=INFO source=runner.go:936 msg="starting go runner" time=2025-02-20T15:49:31.616+08:00 level=INFO source=runner.go:937 msg=system info="CPU : LLAMAFILE = 1 | CPU : LLAMAFILE = 1 | cgo(clang)" threads=6 time=2025-02-20T15:49:31.616+08:00 level=INFO source=runner.go:995 msg="Server listening on 127.0.0.1:50269" ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-alderlake.dll ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-icelake.dll ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-sandybridge.dll ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-skylakex.dll llama_model_loader: loaded meta data with 26 key-value pairs and 579 tensors from D:\ollama\models\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 llm_load_vocab: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect llm_load_vocab: special tokens cache size = 22 llm_load_vocab: token to piece cache size = 0.9310 MB llm_load_print_meta: format = GGUF V3 (latest) llm_load_print_meta: arch = qwen2 llm_load_print_meta: vocab type = BPE llm_load_print_meta: n_vocab = 152064 llm_load_print_meta: n_merges = 151387 llm_load_print_meta: vocab_only = 0 llm_load_print_meta: n_ctx_train = 131072 llm_load_print_meta: n_embd = 5120 llm_load_print_meta: n_layer = 48 llm_load_print_meta: n_head = 40 llm_load_print_meta: n_head_kv = 8 llm_load_print_meta: n_rot = 128 llm_load_print_meta: n_swa = 0 llm_load_print_meta: n_embd_head_k = 128 llm_load_print_meta: n_embd_head_v = 128 llm_load_print_meta: n_gqa = 5 llm_load_print_meta: n_embd_k_gqa = 1024 llm_load_print_meta: n_embd_v_gqa = 1024 llm_load_print_meta: f_norm_eps = 0.0e+00 llm_load_print_meta: f_norm_rms_eps = 1.0e-05 llm_load_print_meta: f_clamp_kqv = 0.0e+00 llm_load_print_meta: f_max_alibi_bias = 0.0e+00 llm_load_print_meta: f_logit_scale = 0.0e+00 llm_load_print_meta: n_ff = 13824 llm_load_print_meta: n_expert = 0 llm_load_print_meta: n_expert_used = 0 llm_load_print_meta: causal attn = 1 llm_load_print_meta: pooling type = 0 llm_load_print_meta: rope type = 2 llm_load_print_meta: rope scaling = linear llm_load_print_meta: freq_base_train = 1000000.0 llm_load_print_meta: freq_scale_train = 1 llm_load_print_meta: n_ctx_orig_yarn = 131072 llm_load_print_meta: rope_finetuned = unknown llm_load_print_meta: ssm_d_conv = 0 llm_load_print_meta: ssm_d_inner = 0 llm_load_print_meta: ssm_d_state = 0 llm_load_print_meta: ssm_dt_rank = 0 llm_load_print_meta: ssm_dt_b_c_rms = 0 llm_load_print_meta: model type = 14B llm_load_print_meta: model ftype = Q4_K - Medium llm_load_print_meta: model params = 14.77 B llm_load_print_meta: model size = 8.37 GiB (4.87 BPW) llm_load_print_meta: general.name = DeepSeek R1 Distill Qwen 14B llm_load_print_meta: BOS token = 151646 '<｜begin▁of▁sentence｜>' llm_load_print_meta: EOS token = 151643 '<｜end▁of▁sentence｜>' llm_load_print_meta: EOT token = 151643 '<｜end▁of▁sentence｜>' llm_load_print_meta: PAD token = 151643 '<｜end▁of▁sentence｜>' llm_load_print_meta: LF token = 148848 'ÄĬ' llm_load_print_meta: FIM PRE token = 151659 '<|fim_prefix|>' llm_load_print_meta: FIM SUF token = 151661 '<|fim_suffix|>' llm_load_print_meta: FIM MID token = 151660 '<|fim_middle|>' llm_load_print_meta: FIM PAD token = 151662 '<|fim_pad|>' llm_load_print_meta: FIM REP token = 151663 '<|repo_name|>' llm_load_print_meta: FIM SEP token = 151664 '<|file_sep|>' llm_load_print_meta: EOG token = 151643 '<｜end▁of▁sentence｜>' llm_load_print_meta: EOG token = 151662 '<|fim_pad|>' llm_load_print_meta: EOG token = 151663 '<|repo_name|>' llm_load_print_meta: EOG token = 151664 '<|file_sep|>' llm_load_print_meta: max token length = 256 llm_load_tensors: CPU model buffer size = 8566.04 MiB time=2025-02-20T15:49:31.845+08:00 level=INFO source=server.go:591 msg="waiting for server to become available" status="llm server loading model" llama_new_context_with_model: n_seq_max = 4 llama_new_context_with_model: n_ctx = 8192 llama_new_context_with_model: n_ctx_per_seq = 2048 llama_new_context_with_model: n_batch = 2048 llama_new_context_with_model: n_ubatch = 512 llama_new_context_with_model: flash_attn = 0 llama_new_context_with_model: freq_base = 1000000.0 llama_new_context_with_model: freq_scale = 1 llama_new_context_with_model: n_ctx_per_seq (2048) < n_ctx_train (131072) -- the full capacity of the model will not be utilized llama_kv_cache_init: kv_size = 8192, offload = 1, type_k = 'f16', type_v = 'f16', n_layer = 48, can_shift = 1 llama_kv_cache_init: CPU KV buffer size = 1536.00 MiB llama_new_context_with_model: KV self size = 1536.00 MiB, K (f16): 768.00 MiB, V (f16): 768.00 MiB llama_new_context_with_model: CPU output buffer size = 2.40 MiB llama_new_context_with_model: CPU compute buffer size = 696.01 MiB llama_new_context_with_model: graph nodes = 1686 llama_new_context_with_model: graph splits = 1 [GIN] 2025/02/20 - 15:49:38 | 200 | 7.1239561s | 127.0.0.1 | POST "/api/generate" time=2025-02-20T15:49:38.605+08:00 level=INFO source=server.go:596 msg="llama runner started in 7.01 seconds" [GIN] 2025/02/20 - 15:49:54 | 200 | 7.687977s | 127.0.0.1 | POST "/api/chat" ``` ### OS Windows ### GPU Nvidia ### CPU Intel ### Ollama version 0.5.11
GiteaMirror added the bug label 2026-04-22 12:33:10 -05:00
GiteaMirror commented 2026-04-22 12:33:11 -05:00
Author
Owner
Copy Link

@Hsq12138 commented on GitHub (Feb 20, 2025):

my gpu is 4080super cuda 12.8

<!-- gh-comment-id:2670737801 --> @Hsq12138 commented on GitHub (Feb 20, 2025): my gpu is 4080super cuda 12.8
GiteaMirror commented 2026-04-22 12:33:12 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Feb 20, 2025):

ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-alderlake.dll
ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll
ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-icelake.dll
ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-sandybridge.dll
ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-skylakex.dll

ollama can't find the backends. Maybe try re-installing.

<!-- gh-comment-id:2671122445 --> @rick-github commented on GitHub (Feb 20, 2025): ``` ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-alderlake.dll ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-icelake.dll ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-sandybridge.dll ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-skylakex.dll ``` ollama can't find the backends. Maybe try re-installing.
GiteaMirror commented 2026-04-22 12:33:12 -05:00
Author
Owner
Copy Link

@Hsq12138 commented on GitHub (Feb 21, 2025):

ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-alderlake.dll
ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll
ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-icelake.dll
ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-sandybridge.dll
ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-skylakex.dll

ollama can't find the backends. Maybe try re-installing.

i tried many times of re installing this problem still exsit

<!-- gh-comment-id:2673378989 --> @Hsq12138 commented on GitHub (Feb 21, 2025): > ``` > ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-alderlake.dll > ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll > ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-icelake.dll > ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-sandybridge.dll > ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-skylakex.dll > ``` > > ollama can't find the backends. Maybe try re-installing. i tried many times of re installing this problem still exsit
GiteaMirror commented 2026-04-22 12:33:12 -05:00
Author
Owner
Copy Link

@lry789 commented on GitHub (Feb 21, 2025):

I have the same problem

<!-- gh-comment-id:2673537878 --> @lry789 commented on GitHub (Feb 21, 2025): I have the same problem
GiteaMirror commented 2026-04-22 12:33:13 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Feb 21, 2025):

Set OLLAMA_DEBUG=1 in the server environment and post the resulting logs.

<!-- gh-comment-id:2673913601 --> @rick-github commented on GitHub (Feb 21, 2025): Set `OLLAMA_DEBUG=1` in the [server environment](https://github.com/ollama/ollama/blob/main/docs/faq.md#how-do-i-configure-ollama-server) and post the resulting logs.
GiteaMirror commented 2026-04-22 12:33:13 -05:00
Author
Owner
Copy Link

@lry789 commented on GitHub (Feb 21, 2025):

-- Logs begin at Thu 2023-11-16 11:05:42 CST. --
2月 21 13:31:11 tc ollama[285585]: CUDA driver version: 12.3
2月 21 13:31:11 tc ollama[285585]: calling cuDeviceGetCount
2月 21 13:31:11 tc ollama[285585]: device count 1
2月 21 13:31:11 tc ollama[285585]: time=2025-02-21T13:31:11.754+08:00 level=DEBUG source=gpu.go:125 msg="detected GPUs" count=1 library=/usr/lib/x86_64-linux-gnu/libcuda.so.545.23.08
2月 21 13:31:11 tc ollama[285585]: [GPU-433dad3d-903e-13f1-25aa-8888bfae0c86] CUDA totalMem 16076 mb
2月 21 13:31:11 tc ollama[285585]: [GPU-433dad3d-903e-13f1-25aa-8888bfae0c86] CUDA freeMem 15811 mb
2月 21 13:31:11 tc ollama[285585]: [GPU-433dad3d-903e-13f1-25aa-8888bfae0c86] Compute Capability 8.9
2月 21 13:31:11 tc ollama[285585]: time=2025-02-21T13:31:11.961+08:00 level=DEBUG source=amd_linux.go:419 msg="amdgpu driver not detected /sys/module/amdgpu"
2月 21 13:31:11 tc ollama[285585]: releasing cuda driver library
2月 21 13:31:11 tc ollama[285585]: time=2025-02-21T13:31:11.961+08:00 level=INFO source=types.go:130 msg="inference compute" id=GPU-433dad3d-903e-13f1-25aa-8888bfae0c86 library=cuda variant=v12 compute=8.9 driver=12.3 name="NVIDIA GeForce RTX 4080" total="15.7 GiB" available="15.4 GiB"


2月 21 13:33:07 tc ollama[285585]: [GIN] 2025/02/21 - 13:33:07 | 200 |     794.342µs |  192.168.30.134 | GET      "/api/tags"
2月 21 13:33:07 tc ollama[285585]: [GIN] 2025/02/21 - 13:33:07 | 200 |      608.09µs |  192.168.30.134 | GET      "/api/tags"
2月 21 13:33:07 tc ollama[285585]: [GIN] 2025/02/21 - 13:33:07 | 200 |      73.143µs |  192.168.30.134 | GET      "/"
2月 21 13:33:10 tc ollama[285585]: time=2025-02-21T13:33:10.934+08:00 level=DEBUG source=gpu.go:391 msg="updating system memory data" before.total="62.7 GiB" before.free="60.5 GiB" before.free_swap="2.0 GiB" now.total="62.7 GiB" now.free="60.4 GiB" now.free_swap="2.0 GiB"
2月 21 13:33:10 tc ollama[285585]: initializing /usr/lib/x86_64-linux-gnu/libcuda.so.545.23.08
2月 21 13:33:10 tc ollama[285585]: dlsym: cuInit - 0x7fc0ae644c30
2月 21 13:33:10 tc ollama[285585]: dlsym: cuDriverGetVersion - 0x7fc0ae644c50
2月 21 13:33:10 tc ollama[285585]: dlsym: cuDeviceGetCount - 0x7fc0ae644c90
2月 21 13:33:10 tc ollama[285585]: dlsym: cuDeviceGet - 0x7fc0ae644c70
2月 21 13:33:10 tc ollama[285585]: dlsym: cuDeviceGetAttribute - 0x7fc0ae644d70
2月 21 13:33:10 tc ollama[285585]: dlsym: cuDeviceGetUuid - 0x7fc0ae644cd0
2月 21 13:33:10 tc ollama[285585]: dlsym: cuDeviceGetName - 0x7fc0ae644cb0
2月 21 13:33:10 tc ollama[285585]: dlsym: cuCtxCreate_v3 - 0x7fc0ae644f50
2月 21 13:33:10 tc ollama[285585]: dlsym: cuMemGetInfo_v2 - 0x7fc0ae64ec40
2月 21 13:33:10 tc ollama[285585]: dlsym: cuCtxDestroy - 0x7fc0ae69e380
2月 21 13:33:10 tc ollama[285585]: calling cuInit
2月 21 13:33:10 tc ollama[285585]: calling cuDriverGetVersion
2月 21 13:33:10 tc ollama[285585]: raw version 0x2efe
2月 21 13:33:10 tc ollama[285585]: CUDA driver version: 12.3
2月 21 13:33:10 tc ollama[285585]: calling cuDeviceGetCount
2月 21 13:33:10 tc ollama[285585]: device count 1
2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.121+08:00 level=DEBUG source=gpu.go:441 msg="updating cuda memory data" gpu=GPU-433dad3d-903e-13f1-25aa-8888bfae0c86 name="NVIDIA GeForce RTX 4080" overhead="0 B" before.total="15.7 GiB" before.free="15.4 GiB" now.total="15.7 GiB" now.free="15.4 GiB" now.used="265.1 MiB"
2月 21 13:33:11 tc ollama[285585]: releasing cuda driver library
2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.164+08:00 level=DEBUG source=sched.go:224 msg="loading first model" model=/home/tc/.ollama/models/blobs/sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e
2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.164+08:00 level=DEBUG source=memory.go:107 msg=evaluating library=cuda gpu_count=1 available="[15.4 GiB]"
2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.166+08:00 level=INFO source=sched.go:714 msg="new model will fit in available VRAM in single GPU, loading" model=/home/tc/.ollama/models/blobs/sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e gpu=GPU-433dad3d-903e-13f1-25aa-8888bfae0c86 parallel=1 available=16579887104 required="9.2 GiB"
2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.166+08:00 level=DEBUG source=gpu.go:391 msg="updating system memory data" before.total="62.7 GiB" before.free="60.4 GiB" before.free_swap="2.0 GiB" now.total="62.7 GiB" now.free="60.4 GiB" now.free_swap="2.0 GiB"
2月 21 13:33:11 tc ollama[285585]: initializing /usr/lib/x86_64-linux-gnu/libcuda.so.545.23.08
2月 21 13:33:11 tc ollama[285585]: dlsym: cuInit - 0x7fc0ae644c30
2月 21 13:33:11 tc ollama[285585]: dlsym: cuDriverGetVersion - 0x7fc0ae644c50
2月 21 13:33:11 tc ollama[285585]: dlsym: cuDeviceGetCount - 0x7fc0ae644c90
2月 21 13:33:11 tc ollama[285585]: dlsym: cuDeviceGet - 0x7fc0ae644c70
2月 21 13:33:11 tc ollama[285585]: dlsym: cuDeviceGetAttribute - 0x7fc0ae644d70
2月 21 13:33:11 tc ollama[285585]: dlsym: cuDeviceGetUuid - 0x7fc0ae644cd0
2月 21 13:33:11 tc ollama[285585]: dlsym: cuDeviceGetName - 0x7fc0ae644cb0
2月 21 13:33:11 tc ollama[285585]: dlsym: cuCtxCreate_v3 - 0x7fc0ae644f50
2月 21 13:33:11 tc ollama[285585]: dlsym: cuMemGetInfo_v2 - 0x7fc0ae64ec40
2月 21 13:33:11 tc ollama[285585]: dlsym: cuCtxDestroy - 0x7fc0ae69e380
2月 21 13:33:11 tc ollama[285585]: calling cuInit
2月 21 13:33:11 tc ollama[285585]: calling cuDriverGetVersion
2月 21 13:33:11 tc ollama[285585]: raw version 0x2efe
2月 21 13:33:11 tc ollama[285585]: CUDA driver version: 12.3
2月 21 13:33:11 tc ollama[285585]: calling cuDeviceGetCount
2月 21 13:33:11 tc ollama[285585]: device count 1
2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.342+08:00 level=DEBUG source=gpu.go:441 msg="updating cuda memory data" gpu=GPU-433dad3d-903e-13f1-25aa-8888bfae0c86 name="NVIDIA GeForce RTX 4080" overhead="0 B" before.total="15.7 GiB" before.free="15.4 GiB" now.total="15.7 GiB" now.free="15.4 GiB" now.used="265.1 MiB"
2月 21 13:33:11 tc ollama[285585]: releasing cuda driver library
2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.342+08:00 level=INFO source=server.go:100 msg="system memory" total="62.7 GiB" free="60.4 GiB" free_swap="2.0 GiB"
2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.342+08:00 level=DEBUG source=memory.go:107 msg=evaluating library=cuda gpu_count=1 available="[15.4 GiB]"
2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.343+08:00 level=INFO source=memory.go:356 msg="offload to cuda" layers.requested=-1 layers.model=49 layers.offload=49 layers.split="" memory.available="[15.4 GiB]" memory.gpu_overhead="0 B" memory.required.full="9.2 GiB" memory.required.partial="9.2 GiB" memory.required.kv="384.0 MiB" memory.required.allocations="[9.2 GiB]" memory.weights.total="7.7 GiB" memory.weights.repeating="7.1 GiB" memory.weights.nonrepeating="609.1 MiB" memory.graph.full="307.0 MiB" memory.graph.partial="916.1 MiB"
2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.343+08:00 level=DEBUG source=server.go:262 msg="compatible gpu libraries" compatible=[]
2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.343+08:00 level=INFO source=server.go:380 msg="starting llama server" cmd="/usr/local/bin/ollama runner --model /home/tc/.ollama/models/blobs/sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e --ctx-size 2048 --batch-size 512 --n-gpu-layers 49 --verbose --threads 28 --parallel 1 --port 45965"
2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.343+08:00 level=DEBUG source=server.go:398 msg=subprocess environment="[PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/snap/bin CUDA_VISIBLE_DEVICES=GPU-433dad3d-903e-13f1-25aa-8888bfae0c86 LD_LIBRARY_PATH=/usr/local/bin]"
2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.344+08:00 level=INFO source=sched.go:449 msg="loaded runners" count=1
2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.344+08:00 level=INFO source=server.go:557 msg="waiting for llama runner to start responding"
2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.344+08:00 level=INFO source=server.go:591 msg="waiting for server to become available" status="llm server error"
2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.387+08:00 level=INFO source=runner.go:936 msg="starting go runner"
2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.387+08:00 level=INFO source=runner.go:937 msg=system info="CPU : LLAMAFILE = 1 | CPU : LLAMAFILE = 1 | cgo(gcc)" threads=28
2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.387+08:00 level=DEBUG source=ggml.go:89 msg="ggml backend load all from path" path=/usr/local/bin
2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.387+08:00 level=INFO source=runner.go:995 msg="Server listening on 127.0.0.1:45965"
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: loaded meta data with 26 key-value pairs and 579 tensors from /home/tc/.ollama/models/blobs/sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e (version GGUF V3 (latest))
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv   0:                       general.architecture str              = qwen2
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv   1:                               general.type str              = model
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv   2:                               general.name str              = DeepSeek R1 Distill Qwen 14B
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv   3:                           general.basename str              = DeepSeek-R1-Distill-Qwen
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv   4:                         general.size_label str              = 14B
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv   5:                          qwen2.block_count u32              = 48
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv   6:                       qwen2.context_length u32              = 131072
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv   7:                     qwen2.embedding_length u32              = 5120
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv   8:                  qwen2.feed_forward_length u32              = 13824
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv   9:                 qwen2.attention.head_count u32              = 40
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv  10:              qwen2.attention.head_count_kv u32              = 8
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv  11:                       qwen2.rope.freq_base f32              = 1000000.000000
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv  12:     qwen2.attention.layer_norm_rms_epsilon f32              = 0.000010
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv  13:                          general.file_type u32              = 15
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv  14:                       tokenizer.ggml.model str              = gpt2
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv  15:                         tokenizer.ggml.pre str              = qwen2
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv  16:                      tokenizer.ggml.tokens arr[str,152064]  = ["!", "\"", "#", "$", "%", "&", "'", ...
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv  17:                  tokenizer.ggml.token_type arr[i32,152064]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv  18:                      tokenizer.ggml.merges arr[str,151387]  = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv  19:                tokenizer.ggml.bos_token_id u32              = 151646
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv  20:                tokenizer.ggml.eos_token_id u32              = 151643
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv  21:            tokenizer.ggml.padding_token_id u32              = 151643
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv  22:               tokenizer.ggml.add_bos_token bool             = true
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv  23:               tokenizer.ggml.add_eos_token bool             = false
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv  24:                    tokenizer.chat_template str              = {% if not add_generation_prompt is de...
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv  25:               general.quantization_version u32              = 2
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - type  f32:  241 tensors
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - type q4_K:  289 tensors
2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - type q6_K:   49 tensors
2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.596+08:00 level=INFO source=server.go:591 msg="waiting for server to become available" status="llm server loading model"
2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151660 '<|fim_middle|>' is not marked as EOG
2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151659 '<|fim_prefix|>' is not marked as EOG
2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151653 '<|vision_end|>' is not marked as EOG
2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151645 '<｜Assistant｜>' is not marked as EOG
2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151644 '<｜User｜>' is not marked as EOG
2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151655 '<|image_pad|>' is not marked as EOG
2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151651 '<|quad_end|>' is not marked as EOG
2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151646 '<｜begin▁of▁sentence｜>' is not marked as EOG
2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151643 '<｜end▁of▁sentence｜>' is not marked as EOG
2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151652 '<|vision_start|>' is not marked as EOG
2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151647 '<|EOT|>' is not marked as EOG
2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151654 '<|vision_pad|>' is not marked as EOG
2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151656 '<|video_pad|>' is not marked as EOG
2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151661 '<|fim_suffix|>' is not marked as EOG
2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151650 '<|quad_start|>' is not marked as EOG
2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: special tokens cache size = 22
2月 21 13:33:12 tc ollama[285585]: llm_load_vocab: token to piece cache size = 0.9310 MB
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: format           = GGUF V3 (latest)
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: arch             = qwen2
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: vocab type       = BPE
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_vocab          = 152064
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_merges         = 151387
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: vocab_only       = 0
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_ctx_train      = 131072
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_embd           = 5120
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_layer          = 48
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_head           = 40
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_head_kv        = 8
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_rot            = 128
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_swa            = 0
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_embd_head_k    = 128
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_embd_head_v    = 128
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_gqa            = 5
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_embd_k_gqa     = 1024
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_embd_v_gqa     = 1024
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: f_norm_eps       = 0.0e+00
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: f_norm_rms_eps   = 1.0e-05
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: f_clamp_kqv      = 0.0e+00
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: f_max_alibi_bias = 0.0e+00
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: f_logit_scale    = 0.0e+00
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_ff             = 13824
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_expert         = 0
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_expert_used    = 0
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: causal attn      = 1
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: pooling type     = 0
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: rope type        = 2
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: rope scaling     = linear
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: freq_base_train  = 1000000.0
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: freq_scale_train = 1
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_ctx_orig_yarn  = 131072
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: rope_finetuned   = unknown
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: ssm_d_conv       = 0
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: ssm_d_inner      = 0
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: ssm_d_state      = 0
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: ssm_dt_rank      = 0
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: ssm_dt_b_c_rms   = 0
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: model type       = 14B
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: model ftype      = Q4_K - Medium
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: model params     = 14.77 B
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: model size       = 8.37 GiB (4.87 BPW)
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: general.name     = DeepSeek R1 Distill Qwen 14B
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: BOS token        = 151646 '<｜begin▁of▁sentence｜>'
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: EOS token        = 151643 '<｜end▁of▁sentence｜>'
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: EOT token        = 151643 '<｜end▁of▁sentence｜>'
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: PAD token        = 151643 '<｜end▁of▁sentence｜>'
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: LF token         = 148848 'ÄĬ'
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: FIM PRE token    = 151659 '<|fim_prefix|>'
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: FIM SUF token    = 151661 '<|fim_suffix|>'
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: FIM MID token    = 151660 '<|fim_middle|>'
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: FIM PAD token    = 151662 '<|fim_pad|>'
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: FIM REP token    = 151663 '<|repo_name|>'
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: FIM SEP token    = 151664 '<|file_sep|>'
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: EOG token        = 151643 '<｜end▁of▁sentence｜>'
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: EOG token        = 151662 '<|fim_pad|>'
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: EOG token        = 151663 '<|repo_name|>'
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: EOG token        = 151664 '<|file_sep|>'
2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: max token length = 256
2月 21 13:33:12 tc ollama[285585]: llm_load_tensors:   CPU_Mapped model buffer size =  8566.04 MiB
2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: n_seq_max     = 1
2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: n_ctx         = 2048
2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: n_ctx_per_seq = 2048
2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: n_batch       = 512
2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: n_ubatch      = 512
2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: flash_attn    = 0
2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: freq_base     = 1000000.0
2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: freq_scale    = 1
2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: n_ctx_per_seq (2048) < n_ctx_train (131072) -- the full capacity of the model will not be utilized
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: kv_size = 2048, offload = 1, type_k = 'f16', type_v = 'f16', n_layer = 48, can_shift = 1
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 0: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 1: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 2: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 3: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 4: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 5: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 6: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 7: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 8: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 9: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 10: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 11: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 12: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 13: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 14: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 15: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 16: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 17: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 18: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 19: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 20: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 21: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 22: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 23: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 24: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 25: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 26: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 27: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 28: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 29: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 30: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 31: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 32: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 33: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 34: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 35: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 36: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 37: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 38: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 39: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 40: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 41: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 42: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 43: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 44: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 45: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 46: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 47: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
2月 21 13:33:12 tc ollama[285585]: time=2025-02-21T13:33:12.601+08:00 level=DEBUG source=server.go:602 msg="model load progress 1.00"
2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init:        CPU KV buffer size =   384.00 MiB
2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: KV self size  =  384.00 MiB, K (f16):  192.00 MiB, V (f16):  192.00 MiB
2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model:        CPU  output buffer size =     0.60 MiB
2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model:        CPU compute buffer size =   307.00 MiB
2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: graph nodes  = 1686
2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: graph splits = 1
2月 21 13:33:12 tc ollama[285585]: time=2025-02-21T13:33:12.852+08:00 level=INFO source=server.go:596 msg="llama runner started in 1.51 seconds"
2月 21 13:33:12 tc ollama[285585]: time=2025-02-21T13:33:12.852+08:00 level=DEBUG source=sched.go:462 msg="finished setting up runner" model=/home/tc/.ollama/models/blobs/sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e
2月 21 13:33:12 tc ollama[285585]: time=2025-02-21T13:33:12.852+08:00 level=DEBUG source=routes.go:1461 msg="chat request" images=0 prompt="<｜User｜>\n你好<｜Assistant｜>"
2月 21 13:33:12 tc ollama[285585]: time=2025-02-21T13:33:12.857+08:00 level=DEBUG source=cache.go:104 msg="loading cache slot" id=0 cache=0 prompt=5 used=0 remaining=5
2月 21 13:33:20 tc ollama[285585]: [GIN] 2025/02/21 - 13:33:20 | 200 |    1.734805ms |  192.168.30.134 | GET      "/api/tags"
2月 21 13:33:20 tc ollama[285585]: [GIN] 2025/02/21 - 13:33:20 | 200 |     508.613µs |  192.168.30.134 | GET      "/api/tags"
2月 21 13:33:22 tc ollama[285585]: [GIN] 2025/02/21 - 13:33:22 | 200 | 11.844414043s |  192.168.30.134 | POST     "/api/chat"
2月 21 13:33:22 tc ollama[285585]: time=2025-02-21T13:33:22.747+08:00 level=DEBUG source=sched.go:466 msg="context for request finished"
2月 21 13:33:22 tc ollama[285585]: time=2025-02-21T13:33:22.747+08:00 level=DEBUG source=sched.go:339 msg="runner with non-zero duration has gone idle, adding timer" modelPath=/home/tc/.ollama/models/blobs/sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e duration=24h0m0s
2月 21 13:33:22 tc ollama[285585]: time=2025-02-21T13:33:22.747+08:00 level=DEBUG source=sched.go:357 msg="after processing request finished event" modelPath=/home/tc/.ollama/models/blobs/sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e refCount=0
2月 21 13:33:23 tc ollama[285585]: [GIN] 2025/02/21 - 13:33:23 | 200 |     664.578µs |  192.168.30.134 | GET      "/api/tags"
2月 21 13:33:23 tc ollama[285585]: [GIN] 2025/02/21 - 13:33:23 | 200 |     572.873µs |  192.168.30.134 | GET      "/api/tags"
2月 21 13:33:31 tc ollama[285585]: [GIN] 2025/02/21 - 13:33:31 | 200 |     560.001µs |  192.168.30.134 | GET      "/api/tags"
2月 21 13:33:31 tc ollama[285585]: [GIN] 2025/02/21 - 13:33:31 | 200 |     467.113µs |  192.168.30.134 | GET      "/api/tags"
2月 21 13:37:58 tc ollama[285585]: [GIN] 2025/02/21 - 13:37:58 | 200 |     761.399µs |  192.168.30.134 | GET      "/api/tags"

[Unit]
Description=Ollama Service
After=network-online.target

[Service]
Environment="OLLAMA_GPU=1"
Environment="OLLAMA_DEBUG=1"
Environment="OLLAMA_NUM_PARALLEL=1"
Environment="OLLAMA_MAX_LOADED_MODELS=1"
Environment="OLLAMA_KEEP_ALIVE=24h"
Environment="OLLAMA_ORIGIN="*""
Environment="OLLAMA_MODELS=/home/hide/.ollama/models"
Environment="OLLAMA_HOST=192.168.30.138:16006"
Environment="CUDA_VISIBLE_DEVICES=0"
Environment="OLLAMA_GPU_LAYERS=100"
ExecStart=/usr/local/bin/ollama serve
User=ollama
Group=ollama
Restart=always
RestartSec=3
Environment="PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/snap/bin"

[Install]
WantedBy=default.target
<!-- gh-comment-id:2674043194 --> @lry789 commented on GitHub (Feb 21, 2025): ``` -- Logs begin at Thu 2023-11-16 11:05:42 CST. -- 2月 21 13:31:11 tc ollama[285585]: CUDA driver version: 12.3 2月 21 13:31:11 tc ollama[285585]: calling cuDeviceGetCount 2月 21 13:31:11 tc ollama[285585]: device count 1 2月 21 13:31:11 tc ollama[285585]: time=2025-02-21T13:31:11.754+08:00 level=DEBUG source=gpu.go:125 msg="detected GPUs" count=1 library=/usr/lib/x86_64-linux-gnu/libcuda.so.545.23.08 2月 21 13:31:11 tc ollama[285585]: [GPU-433dad3d-903e-13f1-25aa-8888bfae0c86] CUDA totalMem 16076 mb 2月 21 13:31:11 tc ollama[285585]: [GPU-433dad3d-903e-13f1-25aa-8888bfae0c86] CUDA freeMem 15811 mb 2月 21 13:31:11 tc ollama[285585]: [GPU-433dad3d-903e-13f1-25aa-8888bfae0c86] Compute Capability 8.9 2月 21 13:31:11 tc ollama[285585]: time=2025-02-21T13:31:11.961+08:00 level=DEBUG source=amd_linux.go:419 msg="amdgpu driver not detected /sys/module/amdgpu" 2月 21 13:31:11 tc ollama[285585]: releasing cuda driver library 2月 21 13:31:11 tc ollama[285585]: time=2025-02-21T13:31:11.961+08:00 level=INFO source=types.go:130 msg="inference compute" id=GPU-433dad3d-903e-13f1-25aa-8888bfae0c86 library=cuda variant=v12 compute=8.9 driver=12.3 name="NVIDIA GeForce RTX 4080" total="15.7 GiB" available="15.4 GiB" 2月 21 13:33:07 tc ollama[285585]: [GIN] 2025/02/21 - 13:33:07 | 200 | 794.342µs | 192.168.30.134 | GET "/api/tags" 2月 21 13:33:07 tc ollama[285585]: [GIN] 2025/02/21 - 13:33:07 | 200 | 608.09µs | 192.168.30.134 | GET "/api/tags" 2月 21 13:33:07 tc ollama[285585]: [GIN] 2025/02/21 - 13:33:07 | 200 | 73.143µs | 192.168.30.134 | GET "/" 2月 21 13:33:10 tc ollama[285585]: time=2025-02-21T13:33:10.934+08:00 level=DEBUG source=gpu.go:391 msg="updating system memory data" before.total="62.7 GiB" before.free="60.5 GiB" before.free_swap="2.0 GiB" now.total="62.7 GiB" now.free="60.4 GiB" now.free_swap="2.0 GiB" 2月 21 13:33:10 tc ollama[285585]: initializing /usr/lib/x86_64-linux-gnu/libcuda.so.545.23.08 2月 21 13:33:10 tc ollama[285585]: dlsym: cuInit - 0x7fc0ae644c30 2月 21 13:33:10 tc ollama[285585]: dlsym: cuDriverGetVersion - 0x7fc0ae644c50 2月 21 13:33:10 tc ollama[285585]: dlsym: cuDeviceGetCount - 0x7fc0ae644c90 2月 21 13:33:10 tc ollama[285585]: dlsym: cuDeviceGet - 0x7fc0ae644c70 2月 21 13:33:10 tc ollama[285585]: dlsym: cuDeviceGetAttribute - 0x7fc0ae644d70 2月 21 13:33:10 tc ollama[285585]: dlsym: cuDeviceGetUuid - 0x7fc0ae644cd0 2月 21 13:33:10 tc ollama[285585]: dlsym: cuDeviceGetName - 0x7fc0ae644cb0 2月 21 13:33:10 tc ollama[285585]: dlsym: cuCtxCreate_v3 - 0x7fc0ae644f50 2月 21 13:33:10 tc ollama[285585]: dlsym: cuMemGetInfo_v2 - 0x7fc0ae64ec40 2月 21 13:33:10 tc ollama[285585]: dlsym: cuCtxDestroy - 0x7fc0ae69e380 2月 21 13:33:10 tc ollama[285585]: calling cuInit 2月 21 13:33:10 tc ollama[285585]: calling cuDriverGetVersion 2月 21 13:33:10 tc ollama[285585]: raw version 0x2efe 2月 21 13:33:10 tc ollama[285585]: CUDA driver version: 12.3 2月 21 13:33:10 tc ollama[285585]: calling cuDeviceGetCount 2月 21 13:33:10 tc ollama[285585]: device count 1 2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.121+08:00 level=DEBUG source=gpu.go:441 msg="updating cuda memory data" gpu=GPU-433dad3d-903e-13f1-25aa-8888bfae0c86 name="NVIDIA GeForce RTX 4080" overhead="0 B" before.total="15.7 GiB" before.free="15.4 GiB" now.total="15.7 GiB" now.free="15.4 GiB" now.used="265.1 MiB" 2月 21 13:33:11 tc ollama[285585]: releasing cuda driver library 2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.164+08:00 level=DEBUG source=sched.go:224 msg="loading first model" model=/home/tc/.ollama/models/blobs/sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e 2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.164+08:00 level=DEBUG source=memory.go:107 msg=evaluating library=cuda gpu_count=1 available="[15.4 GiB]" 2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.166+08:00 level=INFO source=sched.go:714 msg="new model will fit in available VRAM in single GPU, loading" model=/home/tc/.ollama/models/blobs/sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e gpu=GPU-433dad3d-903e-13f1-25aa-8888bfae0c86 parallel=1 available=16579887104 required="9.2 GiB" 2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.166+08:00 level=DEBUG source=gpu.go:391 msg="updating system memory data" before.total="62.7 GiB" before.free="60.4 GiB" before.free_swap="2.0 GiB" now.total="62.7 GiB" now.free="60.4 GiB" now.free_swap="2.0 GiB" 2月 21 13:33:11 tc ollama[285585]: initializing /usr/lib/x86_64-linux-gnu/libcuda.so.545.23.08 2月 21 13:33:11 tc ollama[285585]: dlsym: cuInit - 0x7fc0ae644c30 2月 21 13:33:11 tc ollama[285585]: dlsym: cuDriverGetVersion - 0x7fc0ae644c50 2月 21 13:33:11 tc ollama[285585]: dlsym: cuDeviceGetCount - 0x7fc0ae644c90 2月 21 13:33:11 tc ollama[285585]: dlsym: cuDeviceGet - 0x7fc0ae644c70 2月 21 13:33:11 tc ollama[285585]: dlsym: cuDeviceGetAttribute - 0x7fc0ae644d70 2月 21 13:33:11 tc ollama[285585]: dlsym: cuDeviceGetUuid - 0x7fc0ae644cd0 2月 21 13:33:11 tc ollama[285585]: dlsym: cuDeviceGetName - 0x7fc0ae644cb0 2月 21 13:33:11 tc ollama[285585]: dlsym: cuCtxCreate_v3 - 0x7fc0ae644f50 2月 21 13:33:11 tc ollama[285585]: dlsym: cuMemGetInfo_v2 - 0x7fc0ae64ec40 2月 21 13:33:11 tc ollama[285585]: dlsym: cuCtxDestroy - 0x7fc0ae69e380 2月 21 13:33:11 tc ollama[285585]: calling cuInit 2月 21 13:33:11 tc ollama[285585]: calling cuDriverGetVersion 2月 21 13:33:11 tc ollama[285585]: raw version 0x2efe 2月 21 13:33:11 tc ollama[285585]: CUDA driver version: 12.3 2月 21 13:33:11 tc ollama[285585]: calling cuDeviceGetCount 2月 21 13:33:11 tc ollama[285585]: device count 1 2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.342+08:00 level=DEBUG source=gpu.go:441 msg="updating cuda memory data" gpu=GPU-433dad3d-903e-13f1-25aa-8888bfae0c86 name="NVIDIA GeForce RTX 4080" overhead="0 B" before.total="15.7 GiB" before.free="15.4 GiB" now.total="15.7 GiB" now.free="15.4 GiB" now.used="265.1 MiB" 2月 21 13:33:11 tc ollama[285585]: releasing cuda driver library 2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.342+08:00 level=INFO source=server.go:100 msg="system memory" total="62.7 GiB" free="60.4 GiB" free_swap="2.0 GiB" 2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.342+08:00 level=DEBUG source=memory.go:107 msg=evaluating library=cuda gpu_count=1 available="[15.4 GiB]" 2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.343+08:00 level=INFO source=memory.go:356 msg="offload to cuda" layers.requested=-1 layers.model=49 layers.offload=49 layers.split="" memory.available="[15.4 GiB]" memory.gpu_overhead="0 B" memory.required.full="9.2 GiB" memory.required.partial="9.2 GiB" memory.required.kv="384.0 MiB" memory.required.allocations="[9.2 GiB]" memory.weights.total="7.7 GiB" memory.weights.repeating="7.1 GiB" memory.weights.nonrepeating="609.1 MiB" memory.graph.full="307.0 MiB" memory.graph.partial="916.1 MiB" 2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.343+08:00 level=DEBUG source=server.go:262 msg="compatible gpu libraries" compatible=[] 2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.343+08:00 level=INFO source=server.go:380 msg="starting llama server" cmd="/usr/local/bin/ollama runner --model /home/tc/.ollama/models/blobs/sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e --ctx-size 2048 --batch-size 512 --n-gpu-layers 49 --verbose --threads 28 --parallel 1 --port 45965" 2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.343+08:00 level=DEBUG source=server.go:398 msg=subprocess environment="[PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/snap/bin CUDA_VISIBLE_DEVICES=GPU-433dad3d-903e-13f1-25aa-8888bfae0c86 LD_LIBRARY_PATH=/usr/local/bin]" 2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.344+08:00 level=INFO source=sched.go:449 msg="loaded runners" count=1 2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.344+08:00 level=INFO source=server.go:557 msg="waiting for llama runner to start responding" 2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.344+08:00 level=INFO source=server.go:591 msg="waiting for server to become available" status="llm server error" 2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.387+08:00 level=INFO source=runner.go:936 msg="starting go runner" 2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.387+08:00 level=INFO source=runner.go:937 msg=system info="CPU : LLAMAFILE = 1 | CPU : LLAMAFILE = 1 | cgo(gcc)" threads=28 2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.387+08:00 level=DEBUG source=ggml.go:89 msg="ggml backend load all from path" path=/usr/local/bin 2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.387+08:00 level=INFO source=runner.go:995 msg="Server listening on 127.0.0.1:45965" 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: loaded meta data with 26 key-value pairs and 579 tensors from /home/tc/.ollama/models/blobs/sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e (version GGUF V3 (latest)) 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output. 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 0: general.architecture str = qwen2 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 1: general.type str = model 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 2: general.name str = DeepSeek R1 Distill Qwen 14B 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 3: general.basename str = DeepSeek-R1-Distill-Qwen 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 4: general.size_label str = 14B 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 5: qwen2.block_count u32 = 48 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 6: qwen2.context_length u32 = 131072 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 7: qwen2.embedding_length u32 = 5120 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 8: qwen2.feed_forward_length u32 = 13824 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 9: qwen2.attention.head_count u32 = 40 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 10: qwen2.attention.head_count_kv u32 = 8 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 11: qwen2.rope.freq_base f32 = 1000000.000000 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 12: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 13: general.file_type u32 = 15 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 14: tokenizer.ggml.model str = gpt2 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 15: tokenizer.ggml.pre str = qwen2 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 16: tokenizer.ggml.tokens arr[str,152064] = ["!", "\"", "#", "$", "%", "&", "'", ... 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 18: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 151646 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151643 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = true 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 24: tokenizer.chat_template str = {% if not add_generation_prompt is de... 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - kv 25: general.quantization_version u32 = 2 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - type f32: 241 tensors 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - type q4_K: 289 tensors 2月 21 13:33:11 tc ollama[285585]: llama_model_loader: - type q6_K: 49 tensors 2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.596+08:00 level=INFO source=server.go:591 msg="waiting for server to become available" status="llm server loading model" 2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151660 '<|fim_middle|>' is not marked as EOG 2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151659 '<|fim_prefix|>' is not marked as EOG 2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151653 '<|vision_end|>' is not marked as EOG 2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151645 '<｜Assistant｜>' is not marked as EOG 2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151644 '<｜User｜>' is not marked as EOG 2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151655 '<|image_pad|>' is not marked as EOG 2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151651 '<|quad_end|>' is not marked as EOG 2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151646 '<｜begin▁of▁sentence｜>' is not marked as EOG 2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151643 '<｜end▁of▁sentence｜>' is not marked as EOG 2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151652 '<|vision_start|>' is not marked as EOG 2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151647 '<|EOT|>' is not marked as EOG 2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151654 '<|vision_pad|>' is not marked as EOG 2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151656 '<|video_pad|>' is not marked as EOG 2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151661 '<|fim_suffix|>' is not marked as EOG 2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: control token: 151650 '<|quad_start|>' is not marked as EOG 2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect 2月 21 13:33:11 tc ollama[285585]: llm_load_vocab: special tokens cache size = 22 2月 21 13:33:12 tc ollama[285585]: llm_load_vocab: token to piece cache size = 0.9310 MB 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: format = GGUF V3 (latest) 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: arch = qwen2 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: vocab type = BPE 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_vocab = 152064 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_merges = 151387 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: vocab_only = 0 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_ctx_train = 131072 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_embd = 5120 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_layer = 48 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_head = 40 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_head_kv = 8 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_rot = 128 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_swa = 0 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_embd_head_k = 128 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_embd_head_v = 128 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_gqa = 5 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_embd_k_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: f_norm_eps = 0.0e+00 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: f_norm_rms_eps = 1.0e-05 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: f_clamp_kqv = 0.0e+00 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: f_max_alibi_bias = 0.0e+00 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: f_logit_scale = 0.0e+00 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_ff = 13824 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_expert = 0 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_expert_used = 0 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: causal attn = 1 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: pooling type = 0 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: rope type = 2 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: rope scaling = linear 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: freq_base_train = 1000000.0 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: freq_scale_train = 1 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: n_ctx_orig_yarn = 131072 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: rope_finetuned = unknown 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: ssm_d_conv = 0 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: ssm_d_inner = 0 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: ssm_d_state = 0 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: ssm_dt_rank = 0 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: ssm_dt_b_c_rms = 0 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: model type = 14B 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: model ftype = Q4_K - Medium 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: model params = 14.77 B 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: model size = 8.37 GiB (4.87 BPW) 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: general.name = DeepSeek R1 Distill Qwen 14B 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: BOS token = 151646 '<｜begin▁of▁sentence｜>' 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: EOS token = 151643 '<｜end▁of▁sentence｜>' 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: EOT token = 151643 '<｜end▁of▁sentence｜>' 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: PAD token = 151643 '<｜end▁of▁sentence｜>' 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: LF token = 148848 'ÄĬ' 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: FIM PRE token = 151659 '<|fim_prefix|>' 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: FIM SUF token = 151661 '<|fim_suffix|>' 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: FIM MID token = 151660 '<|fim_middle|>' 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: FIM PAD token = 151662 '<|fim_pad|>' 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: FIM REP token = 151663 '<|repo_name|>' 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: FIM SEP token = 151664 '<|file_sep|>' 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: EOG token = 151643 '<｜end▁of▁sentence｜>' 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: EOG token = 151662 '<|fim_pad|>' 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: EOG token = 151663 '<|repo_name|>' 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: EOG token = 151664 '<|file_sep|>' 2月 21 13:33:12 tc ollama[285585]: llm_load_print_meta: max token length = 256 2月 21 13:33:12 tc ollama[285585]: llm_load_tensors: CPU_Mapped model buffer size = 8566.04 MiB 2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: n_seq_max = 1 2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: n_ctx = 2048 2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: n_ctx_per_seq = 2048 2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: n_batch = 512 2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: n_ubatch = 512 2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: flash_attn = 0 2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: freq_base = 1000000.0 2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: freq_scale = 1 2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: n_ctx_per_seq (2048) < n_ctx_train (131072) -- the full capacity of the model will not be utilized 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: kv_size = 2048, offload = 1, type_k = 'f16', type_v = 'f16', n_layer = 48, can_shift = 1 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 0: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 1: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 2: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 3: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 4: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 5: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 6: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 7: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 8: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 9: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 10: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 11: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 12: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 13: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 14: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 15: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 16: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 17: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 18: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 19: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 20: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 21: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 22: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 23: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 24: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 25: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 26: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 27: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 28: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 29: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 30: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 31: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 32: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 33: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 34: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 35: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 36: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 37: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 38: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 39: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 40: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 41: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 42: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 43: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 44: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 45: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 46: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: layer 47: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 2月 21 13:33:12 tc ollama[285585]: time=2025-02-21T13:33:12.601+08:00 level=DEBUG source=server.go:602 msg="model load progress 1.00" 2月 21 13:33:12 tc ollama[285585]: llama_kv_cache_init: CPU KV buffer size = 384.00 MiB 2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: KV self size = 384.00 MiB, K (f16): 192.00 MiB, V (f16): 192.00 MiB 2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: CPU output buffer size = 0.60 MiB 2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: CPU compute buffer size = 307.00 MiB 2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: graph nodes = 1686 2月 21 13:33:12 tc ollama[285585]: llama_new_context_with_model: graph splits = 1 2月 21 13:33:12 tc ollama[285585]: time=2025-02-21T13:33:12.852+08:00 level=INFO source=server.go:596 msg="llama runner started in 1.51 seconds" 2月 21 13:33:12 tc ollama[285585]: time=2025-02-21T13:33:12.852+08:00 level=DEBUG source=sched.go:462 msg="finished setting up runner" model=/home/tc/.ollama/models/blobs/sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e 2月 21 13:33:12 tc ollama[285585]: time=2025-02-21T13:33:12.852+08:00 level=DEBUG source=routes.go:1461 msg="chat request" images=0 prompt="<｜User｜>\n你好<｜Assistant｜>" 2月 21 13:33:12 tc ollama[285585]: time=2025-02-21T13:33:12.857+08:00 level=DEBUG source=cache.go:104 msg="loading cache slot" id=0 cache=0 prompt=5 used=0 remaining=5 2月 21 13:33:20 tc ollama[285585]: [GIN] 2025/02/21 - 13:33:20 | 200 | 1.734805ms | 192.168.30.134 | GET "/api/tags" 2月 21 13:33:20 tc ollama[285585]: [GIN] 2025/02/21 - 13:33:20 | 200 | 508.613µs | 192.168.30.134 | GET "/api/tags" 2月 21 13:33:22 tc ollama[285585]: [GIN] 2025/02/21 - 13:33:22 | 200 | 11.844414043s | 192.168.30.134 | POST "/api/chat" 2月 21 13:33:22 tc ollama[285585]: time=2025-02-21T13:33:22.747+08:00 level=DEBUG source=sched.go:466 msg="context for request finished" 2月 21 13:33:22 tc ollama[285585]: time=2025-02-21T13:33:22.747+08:00 level=DEBUG source=sched.go:339 msg="runner with non-zero duration has gone idle, adding timer" modelPath=/home/tc/.ollama/models/blobs/sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e duration=24h0m0s 2月 21 13:33:22 tc ollama[285585]: time=2025-02-21T13:33:22.747+08:00 level=DEBUG source=sched.go:357 msg="after processing request finished event" modelPath=/home/tc/.ollama/models/blobs/sha256-6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e refCount=0 2月 21 13:33:23 tc ollama[285585]: [GIN] 2025/02/21 - 13:33:23 | 200 | 664.578µs | 192.168.30.134 | GET "/api/tags" 2月 21 13:33:23 tc ollama[285585]: [GIN] 2025/02/21 - 13:33:23 | 200 | 572.873µs | 192.168.30.134 | GET "/api/tags" 2月 21 13:33:31 tc ollama[285585]: [GIN] 2025/02/21 - 13:33:31 | 200 | 560.001µs | 192.168.30.134 | GET "/api/tags" 2月 21 13:33:31 tc ollama[285585]: [GIN] 2025/02/21 - 13:33:31 | 200 | 467.113µs | 192.168.30.134 | GET "/api/tags" 2月 21 13:37:58 tc ollama[285585]: [GIN] 2025/02/21 - 13:37:58 | 200 | 761.399µs | 192.168.30.134 | GET "/api/tags" ``` --------------------------- ``` [Unit] Description=Ollama Service After=network-online.target [Service] Environment="OLLAMA_GPU=1" Environment="OLLAMA_DEBUG=1" Environment="OLLAMA_NUM_PARALLEL=1" Environment="OLLAMA_MAX_LOADED_MODELS=1" Environment="OLLAMA_KEEP_ALIVE=24h" Environment="OLLAMA_ORIGIN="*"" Environment="OLLAMA_MODELS=/home/hide/.ollama/models" Environment="OLLAMA_HOST=192.168.30.138:16006" Environment="CUDA_VISIBLE_DEVICES=0" Environment="OLLAMA_GPU_LAYERS=100" ExecStart=/usr/local/bin/ollama serve User=ollama Group=ollama Restart=always RestartSec=3 Environment="PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/snap/bin" [Install] WantedBy=default.target ```
GiteaMirror commented 2026-04-22 12:33:13 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Feb 21, 2025):

OLLAMA_GPU_LAYERS is not an ollama environment variable.

2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.387+08:00 level=DEBUG source=ggml.go:89 msg="ggml backend load all from path" path=/usr/local/bin

ollama couldn't find an backends to load. What's the output of:

ls -l /usr/local/lib /usr/local/lib/ollama

How did you install ollama?

<!-- gh-comment-id:2674083212 --> @rick-github commented on GitHub (Feb 21, 2025): `OLLAMA_GPU_LAYERS` is not an ollama environment variable. ``` 2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.387+08:00 level=DEBUG source=ggml.go:89 msg="ggml backend load all from path" path=/usr/local/bin ``` ollama couldn't find an backends to load. What's the output of: ``` ls -l /usr/local/lib /usr/local/lib/ollama ``` How did you install ollama?
GiteaMirror commented 2026-04-22 12:33:13 -05:00
Author
Owner
Copy Link

@Hsq12138 commented on GitHub (Feb 21, 2025):

OLLAMA_GPU_LAYERS is not an ollama environment variable.

2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.387+08:00 level=DEBUG source=ggml.go:89 msg="ggml backend load all from path" path=/usr/local/bin

ollama couldn't find an backends to load. What's the output of:

ls -l /usr/local/lib /usr/local/lib/ollama

How did you install ollama?

2025/02/21 23:08:26 routes.go:1187: INFO server config env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_DEBUG:true OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 OLLAMA_INTEL_GPU:false 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:D:\ollama\models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:0 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://] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES:]"
time=2025-02-21T23:08:26.801+08:00 level=INFO source=images.go:432 msg="total blobs: 2"
time=2025-02-21T23:08:26.802+08:00 level=INFO source=images.go:439 msg="total unused blobs removed: 0"
time=2025-02-21T23:08:26.802+08:00 level=INFO source=routes.go:1238 msg="Listening on 127.0.0.1:11434 (version 0.5.12-rc1)"
time=2025-02-21T23:08:26.802+08:00 level=DEBUG source=sched.go:106 msg="starting llm scheduler"
time=2025-02-21T23:08:26.803+08:00 level=INFO source=gpu.go:217 msg="looking for compatible GPUs"
time=2025-02-21T23:08:26.803+08:00 level=INFO source=gpu_windows.go:167 msg=packages count=1
time=2025-02-21T23:08:26.803+08:00 level=INFO source=gpu_windows.go:214 msg="" package=0 cores=6 efficiency=0 threads=12
time=2025-02-21T23:08:26.803+08:00 level=DEBUG source=gpu.go:98 msg="searching for GPU discovery libraries for NVIDIA"
time=2025-02-21T23:08:26.803+08:00 level=DEBUG source=gpu.go:501 msg="Searching for GPU library" name=nvml.dll
time=2025-02-21T23:08:26.803+08:00 level=DEBUG source=gpu.go:525 msg="gpu library search" globs="[C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\nvml.dll C:\Program Files\NVIDIA\CUDNN\v9.7\bin\nvml.dll C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\bin\nvml.dll C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\libnvvp\nvml.dll C:\Users\zrway\AppData\Local\Programs\Ollama\nvml.dll C:\Program Files\Common Files\Oracle\Java\javapath\nvml.dll C:\Windows\system32\nvml.dll C:\Windows\nvml.dll C:\Windows\System32\Wbem\nvml.dll C:\Windows\System32\WindowsPowerShell\v1.0\nvml.dll C:\Windows\System32\OpenSSH\nvml.dll C:\Program Files (x86)\NVIDIA Corporation\PhysX\Common\nvml.dll C:\Program Files\Bandizip\nvml.dll C:\Program Files\dotnet\nvml.dll C:\Program Files\Git\cmd\nvml.dll D:\ai\ffmpeg-2024-03-28-git-5d71f97e0e-full_build\bin\nvml.dll C:\Program Files\NVIDIA Corporation\NVIDIA app\NvDLISR\nvml.dll C:\Users\zrway\AppData\Local\Microsoft\WindowsApps\python.exe\nvml.dll C:\Users\zrway\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.11_qbz5n2kfra8p0\LocalCache\local-packages\Python311\Scripts\nvml.dll C:\Program Files\NVIDIA Corporation\Nsight Compute 2025.1.0\nvml.dll C:\Program Files\MySQL\MySQL Shell 8.0\bin\nvml.dll C:\Users\zrway\AppData\Local\Microsoft\WindowsApps\nvml.dll C:\Users\zrway\AppData\Local\Programs\Microsoft VS Code\bin\nvml.dll D:\ai\ffmpeg-2024-03-28-git-5d71f97e0e-essentials_build\bin\nvml.dll C:\Users\zrway\AppData\Local\Programs\Ollama\nvml.dll C:\Users\zrway\AppData\Local\Programs\Ollama\nvml.dll C:\Users\zrway\.lmstudio\bin\nvml.dll c:\Windows\System32\nvml.dll]"
time=2025-02-21T23:08:26.804+08:00 level=DEBUG source=gpu.go:529 msg="skipping PhysX cuda library path" path="C:\Program Files (x86)\NVIDIA Corporation\PhysX\Common\nvml.dll"
time=2025-02-21T23:08:26.805+08:00 level=DEBUG source=gpu.go:558 msg="discovered GPU libraries" paths="[C:\Windows\system32\nvml.dll c:\Windows\System32\nvml.dll]"
time=2025-02-21T23:08:26.839+08:00 level=DEBUG source=gpu.go:111 msg="nvidia-ml loaded" library=C:\Windows\system32\nvml.dll
time=2025-02-21T23:08:26.839+08:00 level=DEBUG source=gpu.go:501 msg="Searching for GPU library" name=nvcuda.dll
time=2025-02-21T23:08:26.839+08:00 level=DEBUG source=gpu.go:525 msg="gpu library search" globs="[C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\nvcuda.dll C:\Program Files\NVIDIA\CUDNN\v9.7\bin\nvcuda.dll C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\bin\nvcuda.dll C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\libnvvp\nvcuda.dll C:\Users\zrway\AppData\Local\Programs\Ollama\nvcuda.dll C:\Program Files\Common Files\Oracle\Java\javapath\nvcuda.dll C:\Windows\system32\nvcuda.dll C:\Windows\nvcuda.dll C:\Windows\System32\Wbem\nvcuda.dll C:\Windows\System32\WindowsPowerShell\v1.0\nvcuda.dll C:\Windows\System32\OpenSSH\nvcuda.dll C:\Program Files (x86)\NVIDIA Corporation\PhysX\Common\nvcuda.dll C:\Program Files\Bandizip\nvcuda.dll C:\Program Files\dotnet\nvcuda.dll C:\Program Files\Git\cmd\nvcuda.dll D:\ai\ffmpeg-2024-03-28-git-5d71f97e0e-full_build\bin\nvcuda.dll C:\Program Files\NVIDIA Corporation\NVIDIA app\NvDLISR\nvcuda.dll C:\Users\zrway\AppData\Local\Microsoft\WindowsApps\python.exe\nvcuda.dll C:\Users\zrway\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.11_qbz5n2kfra8p0\LocalCache\local-packages\Python311\Scripts\nvcuda.dll C:\Program Files\NVIDIA Corporation\Nsight Compute 2025.1.0\nvcuda.dll C:\Program Files\MySQL\MySQL Shell 8.0\bin\nvcuda.dll C:\Users\zrway\AppData\Local\Microsoft\WindowsApps\nvcuda.dll C:\Users\zrway\AppData\Local\Programs\Microsoft VS Code\bin\nvcuda.dll D:\ai\ffmpeg-2024-03-28-git-5d71f97e0e-essentials_build\bin\nvcuda.dll C:\Users\zrway\AppData\Local\Programs\Ollama\nvcuda.dll C:\Users\zrway\AppData\Local\Programs\Ollama\nvcuda.dll C:\Users\zrway\.lmstudio\bin\nvcuda.dll c:\windows\system\nvcuda.dll]"
time=2025-02-21T23:08:26.840+08:00 level=DEBUG source=gpu.go:529 msg="skipping PhysX cuda library path" path="C:\Program Files (x86)\NVIDIA Corporation\PhysX\Common\nvcuda.dll"
time=2025-02-21T23:08:26.841+08:00 level=DEBUG source=gpu.go:558 msg="discovered GPU libraries" paths=[C:\Windows\system32\nvcuda.dll]
initializing C:\Windows\system32\nvcuda.dll
dlsym: cuInit - 00007FF97B775F80
dlsym: cuDriverGetVersion - 00007FF97B776020
dlsym: cuDeviceGetCount - 00007FF97B776816
dlsym: cuDeviceGet - 00007FF97B776810
dlsym: cuDeviceGetAttribute - 00007FF97B776170
dlsym: cuDeviceGetUuid - 00007FF97B776822
dlsym: cuDeviceGetName - 00007FF97B77681C
dlsym: cuCtxCreate_v3 - 00007FF97B776894
dlsym: cuMemGetInfo_v2 - 00007FF97B776996
dlsym: cuCtxDestroy - 00007FF97B7768A6
calling cuInit
calling cuDriverGetVersion
raw version 0x2f30
CUDA driver version: 12.8
calling cuDeviceGetCount
device count 1
time=2025-02-21T23:08:26.893+08:00 level=DEBUG source=gpu.go:125 msg="detected GPUs" count=1 library=C:\Windows\system32\nvcuda.dll
[GPU-ecc0382b-7d7c-7b61-8572-a21b10ac9fcd] CUDA totalMem 16375 mb
[GPU-ecc0382b-7d7c-7b61-8572-a21b10ac9fcd] CUDA freeMem 15035 mb
[GPU-ecc0382b-7d7c-7b61-8572-a21b10ac9fcd] Compute Capability 8.9
time=2025-02-21T23:08:26.975+08:00 level=INFO source=gpu.go:319 msg="detected OS VRAM overhead" id=GPU-ecc0382b-7d7c-7b61-8572-a21b10ac9fcd library=cuda compute=8.9 driver=12.8 name="NVIDIA GeForce RTX 4080 SUPER" overhead="265.6 MiB"
time=2025-02-21T23:08:26.976+08:00 level=DEBUG source=amd_windows.go:34 msg="unable to load amdhip64_6.dll, please make sure to upgrade to the latest amd driver: The file cannot be accessed by the system."
releasing cuda driver library
releasing nvml library
time=2025-02-21T23:08:26.977+08:00 level=INFO source=types.go:130 msg="inference compute" id=GPU-ecc0382b-7d7c-7b61-8572-a21b10ac9fcd library=cuda variant=v12 compute=8.9 driver=12.8 name="NVIDIA GeForce RTX 4080 SUPER" total="16.0 GiB" available="14.7 GiB"
[GIN] 2025/02/21 - 23:18:16 | 200 | 0s | 127.0.0.1 | HEAD "/"
[GIN] 2025/02/21 - 23:18:16 | 200 | 1.0301ms | 127.0.0.1 | GET "/api/tags"
[GIN] 2025/02/21 - 23:18:31 | 200 | 0s | 127.0.0.1 | HEAD "/"
[GIN] 2025/02/21 - 23:18:31 | 200 | 10.8992ms | 127.0.0.1 | POST "/api/show"
time=2025-02-21T23:18:31.919+08:00 level=DEBUG source=gpu.go:391 msg="updating system memory data" before.total="95.8 GiB" before.free="87.3 GiB" before.free_swap="89.6 GiB" now.total="95.8 GiB" now.free="86.2 GiB" now.free_swap="87.3 GiB"
time=2025-02-21T23:18:31.928+08:00 level=DEBUG source=gpu.go:441 msg="updating cuda memory data" gpu=GPU-ecc0382b-7d7c-7b61-8572-a21b10ac9fcd name="NVIDIA GeForce RTX 4080 SUPER" overhead="265.6 MiB" before.total="16.0 GiB" before.free="14.7 GiB" now.total="16.0 GiB" now.free="14.4 GiB" now.used="1.3 GiB"
releasing nvml library
time=2025-02-21T23:18:31.928+08:00 level=DEBUG source=sched.go:182 msg="updating default concurrency" OLLAMA_MAX_LOADED_MODELS=3 gpu_count=1
time=2025-02-21T23:18:31.946+08:00 level=DEBUG source=sched.go:225 msg="loading first model" model=D:\ollama\models\blobs\sha256-553aa261cfb6856c595c9fefdb5453b98fdef331bf2ca918a5e0a23aa254d022
time=2025-02-21T23:18:31.946+08:00 level=DEBUG source=memory.go:108 msg=evaluating library=cuda gpu_count=1 available="[14.4 GiB]"
time=2025-02-21T23:18:31.946+08:00 level=WARN source=ggml.go:132 msg="key not found" key=qwen2.attention.key_length default=128
time=2025-02-21T23:18:31.946+08:00 level=WARN source=ggml.go:132 msg="key not found" key=qwen2.attention.value_length default=128
time=2025-02-21T23:18:31.947+08:00 level=DEBUG source=memory.go:108 msg=evaluating library=cuda gpu_count=1 available="[14.4 GiB]"
time=2025-02-21T23:18:31.947+08:00 level=WARN source=ggml.go:132 msg="key not found" key=qwen2.attention.key_length default=128
time=2025-02-21T23:18:31.947+08:00 level=WARN source=ggml.go:132 msg="key not found" key=qwen2.attention.value_length default=128
time=2025-02-21T23:18:31.948+08:00 level=DEBUG source=memory.go:108 msg=evaluating library=cuda gpu_count=1 available="[14.4 GiB]"
time=2025-02-21T23:18:31.948+08:00 level=WARN source=ggml.go:132 msg="key not found" key=qwen2.attention.key_length default=128
time=2025-02-21T23:18:31.948+08:00 level=WARN source=ggml.go:132 msg="key not found" key=qwen2.attention.value_length default=128
time=2025-02-21T23:18:31.948+08:00 level=DEBUG source=memory.go:108 msg=evaluating library=cuda gpu_count=1 available="[14.4 GiB]"
time=2025-02-21T23:18:31.948+08:00 level=WARN source=ggml.go:132 msg="key not found" key=qwen2.attention.key_length default=128
time=2025-02-21T23:18:31.948+08:00 level=WARN source=ggml.go:132 msg="key not found" key=qwen2.attention.value_length default=128
time=2025-02-21T23:18:31.948+08:00 level=DEBUG source=gpu.go:391 msg="updating system memory data" before.total="95.8 GiB" before.free="86.2 GiB" before.free_swap="87.3 GiB" now.total="95.8 GiB" now.free="86.2 GiB" now.free_swap="87.3 GiB"
time=2025-02-21T23:18:31.956+08:00 level=DEBUG source=gpu.go:441 msg="updating cuda memory data" gpu=GPU-ecc0382b-7d7c-7b61-8572-a21b10ac9fcd name="NVIDIA GeForce RTX 4080 SUPER" overhead="265.6 MiB" before.total="16.0 GiB" before.free="14.4 GiB" now.total="16.0 GiB" now.free="14.4 GiB" now.used="1.3 GiB"
releasing nvml library
time=2025-02-21T23:18:31.956+08:00 level=INFO source=server.go:97 msg="system memory" total="95.8 GiB" free="86.2 GiB" free_swap="87.3 GiB"
time=2025-02-21T23:18:31.956+08:00 level=DEBUG source=memory.go:108 msg=evaluating library=cuda gpu_count=1 available="[14.4 GiB]"
time=2025-02-21T23:18:31.956+08:00 level=WARN source=ggml.go:132 msg="key not found" key=qwen2.attention.key_length default=128
time=2025-02-21T23:18:31.956+08:00 level=WARN source=ggml.go:132 msg="key not found" key=qwen2.attention.value_length default=128
time=2025-02-21T23:18:31.957+08:00 level=INFO source=server.go:130 msg=offload library=cuda layers.requested=-1 layers.model=49 layers.offload=24 layers.split="" memory.available="[14.4 GiB]" memory.gpu_overhead="0 B" memory.required.full="28.3 GiB" memory.required.partial="14.4 GiB" memory.required.kv="384.0 MiB" memory.required.allocations="[14.4 GiB]" memory.weights.total="25.0 GiB" memory.weights.repeating="23.5 GiB" memory.weights.nonrepeating="1.5 GiB" memory.graph.full="307.0 MiB" memory.graph.partial="916.1 MiB"
time=2025-02-21T23:18:31.957+08:00 level=DEBUG source=server.go:259 msg="compatible gpu libraries" compatible="[cuda_v12 cuda_v11]"
time=2025-02-21T23:18:31.963+08:00 level=DEBUG source=server.go:302 msg="adding gpu library" path=C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\cuda_v12
time=2025-02-21T23:18:31.963+08:00 level=DEBUG source=server.go:310 msg="adding gpu dependency paths" paths=[C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\cuda_v12]
time=2025-02-21T23:18:31.963+08:00 level=INFO source=server.go:380 msg="starting llama server" cmd="C:\Users\zrway\AppData\Local\Programs\Ollama\ollama.exe runner --model D:\ollama\models\blobs\sha256-553aa261cfb6856c595c9fefdb5453b98fdef331bf2ca918a5e0a23aa254d022 --ctx-size 2048 --batch-size 512 --n-gpu-layers 24 --verbose --threads 6 --no-mmap --parallel 1 --port 50841"
time=2025-02-21T23:18:31.963+08:00 level=DEBUG source=server.go:398 msg=subprocess environment="[CUDA_PATH=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8 CUDA_PATH_V12_8=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8 PATH=C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\cuda_v12;C:\Program Files\NVIDIA\CUDNN\v9.7\bin;C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\bin;C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\libnvvp;;C:\Program Files\Common Files\Oracle\Java\javapath;C:\Windows\system32;C:\Windows;C:\Windows\System32\Wbem;C:\Windows\System32\WindowsPowerShell\v1.0\;C:\Windows\System32\OpenSSH\;C:\Program Files (x86)\NVIDIA Corporation\PhysX\Common;C:\Program Files\Bandizip\;C:\Program Files\dotnet\;C:\Program Files\Git\cmd;D:\ai\ffmpeg-2024-03-28-git-5d71f97e0e-full_build\bin;C:\Program Files\NVIDIA Corporation\NVIDIA app\NvDLISR;C:\Users\zrway\AppData\Local\Microsoft\WindowsApps\python.exe;C:\Users\zrway\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.11_qbz5n2kfra8p0\LocalCache\local-packages\Python311\Scripts;C:\Program Files\NVIDIA Corporation\Nsight Compute 2025.1.0\;C:\Program Files\MySQL\MySQL Shell 8.0\bin\;C:\Users\zrway\AppData\Local\Microsoft\WindowsApps;C:\Users\zrway\AppData\Local\Programs\Microsoft VS Code\bin;D:\ai\ffmpeg-2024-03-28-git-5d71f97e0e-essentials_build\bin;;C:\Users\zrway\AppData\Local\Programs\Ollama;C:\Users\zrway\.lmstudio\bin;C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\cuda_v12;C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama CUDA_VISIBLE_DEVICES=GPU-ecc0382b-7d7c-7b61-8572-a21b10ac9fcd]"
time=2025-02-21T23:18:32.020+08:00 level=INFO source=sched.go:450 msg="loaded runners" count=1
time=2025-02-21T23:18:32.020+08:00 level=INFO source=server.go:557 msg="waiting for llama runner to start responding"
time=2025-02-21T23:18:32.020+08:00 level=INFO source=server.go:591 msg="waiting for server to become available" status="llm server error"
time=2025-02-21T23:18:32.035+08:00 level=INFO source=runner.go:932 msg="starting go runner"
time=2025-02-21T23:18:32.041+08:00 level=DEBUG source=ggml.go:89 msg="ggml backend load all from path" path=C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\cuda_v12
time=2025-02-21T23:18:32.051+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path="C:\Program Files\NVIDIA\CUDNN\v9.7\bin"
time=2025-02-21T23:18:32.051+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\bin"
time=2025-02-21T23:18:32.051+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\libnvvp"
time=2025-02-21T23:18:32.051+08:00 level=DEBUG source=ggml.go:89 msg="ggml backend load all from path" path=C:\Users\zrway\AppData\Local\Programs\Ollama
time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path="C:\Program Files\Common Files\Oracle\Java\javapath"
time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path=C:\Windows\system32
time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path=C:\Windows
time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path=C:\Windows\System32\Wbem
time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path=C:\Windows\System32\WindowsPowerShell\v1.0
time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path=C:\Windows\System32\OpenSSH
time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path="C:\Program Files (x86)\NVIDIA Corporation\PhysX\Common"
time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path="C:\Program Files\Bandizip"
time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path="C:\Program Files\dotnet"
time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path="C:\Program Files\Git\cmd"
time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path=D:\ai\ffmpeg-2024-03-28-git-5d71f97e0e-full_build\bin
time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path="C:\Program Files\NVIDIA Corporation\NVIDIA app\NvDLISR"
time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path=C:\Users\zrway\AppData\Local\Microsoft\WindowsApps\python.exe
time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path=C:\Users\zrway\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.11_qbz5n2kfra8p0\LocalCache\local-packages\Python311\Scripts
time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path="C:\Program Files\NVIDIA Corporation\Nsight Compute 2025.1.0"
time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path="C:\Program Files\MySQL\MySQL Shell 8.0\bin"
time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path=C:\Users\zrway\AppData\Local\Microsoft\WindowsApps
time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path="C:\Users\zrway\AppData\Local\Programs\Microsoft VS Code\bin"
time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path=D:\ai\ffmpeg-2024-03-28-git-5d71f97e0e-essentials_build\bin
time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path=C:\Users\zrway.lmstudio\bin
time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:89 msg="ggml backend load all from path" path=C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama
ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-alderlake.dll
ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll
ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-icelake.dll
ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-sandybridge.dll
ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-skylakex.dll
time=2025-02-21T23:18:32.060+08:00 level=INFO source=runner.go:935 msg=system info="CPU : LLAMAFILE = 1 | CPU : LLAMAFILE = 1 | cgo(clang)" threads=6
time=2025-02-21T23:18:32.074+08:00 level=INFO source=runner.go:993 msg="Server listening on 127.0.0.1:50841"
llama_model_loader: loaded meta data with 25 key-value pairs and 579 tensors from D:\ollama\models\blobs\sha256-553aa261cfb6856c595c9fefdb5453b98fdef331bf2ca918a5e0a23aa254d022 (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 = Checkpoint 887 Merged
llama_model_loader: - kv 3: general.size_label str = 15B
llama_model_loader: - kv 4: qwen2.block_count u32 = 48
llama_model_loader: - kv 5: qwen2.context_length u32 = 131072
llama_model_loader: - kv 6: qwen2.embedding_length u32 = 5120
llama_model_loader: - kv 7: qwen2.feed_forward_length u32 = 13824
llama_model_loader: - kv 8: qwen2.attention.head_count u32 = 40
llama_model_loader: - kv 9: qwen2.attention.head_count_kv u32 = 8
llama_model_loader: - kv 10: qwen2.rope.freq_base f32 = 1000000.000000
llama_model_loader: - kv 11: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010
llama_model_loader: - kv 12: general.file_type u32 = 1
llama_model_loader: - kv 13: tokenizer.ggml.model str = gpt2
llama_model_loader: - kv 14: tokenizer.ggml.pre str = deepseek-r1-qwen
llama_model_loader: - kv 15: tokenizer.ggml.tokens arr[str,152064] = ["!", """, "#", "$", "%", "&", "'", ...
llama_model_loader: - kv 16: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv 17: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv 18: tokenizer.ggml.bos_token_id u32 = 151646
llama_model_loader: - kv 19: tokenizer.ggml.eos_token_id u32 = 151643
llama_model_loader: - kv 20: tokenizer.ggml.padding_token_id u32 = 151643
llama_model_loader: - kv 21: tokenizer.ggml.add_bos_token bool = true
llama_model_loader: - kv 22: tokenizer.ggml.add_eos_token bool = false
llama_model_loader: - kv 23: tokenizer.chat_template str = {% if not add_generation_prompt is de...
llama_model_loader: - kv 24: general.quantization_version u32 = 2
llama_model_loader: - type f32: 241 tensors
llama_model_loader: - type f16: 338 tensors
llm_load_vocab: missing or unrecognized pre-tokenizer type, using: 'default'
llm_load_vocab: control token: 151659 '<|fim_prefix|>' is not marked as EOG
llm_load_vocab: control token: 151656 '<|video_pad|>' is not marked as EOG
llm_load_vocab: control token: 151655 '<|image_pad|>' is not marked as EOG
llm_load_vocab: control token: 151653 '<|vision_end|>' is not marked as EOG
llm_load_vocab: control token: 151652 '<|vision_start|>' is not marked as EOG
llm_load_vocab: control token: 151651 '<|quad_end|>' is not marked as EOG
llm_load_vocab: control token: 151646 '<｜begin▁of▁sentence｜>' is not marked as EOG
llm_load_vocab: control token: 151644 '<｜User｜>' is not marked as EOG
llm_load_vocab: control token: 151661 '<|fim_suffix|>' is not marked as EOG
llm_load_vocab: control token: 151660 '<|fim_middle|>' is not marked as EOG
llm_load_vocab: control token: 151654 '<|vision_pad|>' is not marked as EOG
llm_load_vocab: control token: 151650 '<|quad_start|>' is not marked as EOG
llm_load_vocab: control token: 151647 '<|EOT|>' is not marked as EOG
llm_load_vocab: control token: 151643 '<｜end▁of▁sentence｜>' is not marked as EOG
llm_load_vocab: control token: 151645 '<｜Assistant｜>' is not marked as EOG
llm_load_vocab: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
llm_load_vocab: special tokens cache size = 22
llm_load_vocab: token to piece cache size = 0.9310 MB
llm_load_print_meta: format = GGUF V3 (latest)
llm_load_print_meta: arch = qwen2
llm_load_print_meta: vocab type = BPE
llm_load_print_meta: n_vocab = 152064
llm_load_print_meta: n_merges = 151387
llm_load_print_meta: vocab_only = 0
llm_load_print_meta: n_ctx_train = 131072
llm_load_print_meta: n_embd = 5120
llm_load_print_meta: n_layer = 48
llm_load_print_meta: n_head = 40
llm_load_print_meta: n_head_kv = 8
llm_load_print_meta: n_rot = 128
llm_load_print_meta: n_swa = 0
llm_load_print_meta: n_embd_head_k = 128
llm_load_print_meta: n_embd_head_v = 128
llm_load_print_meta: n_gqa = 5
llm_load_print_meta: n_embd_k_gqa = 1024
llm_load_print_meta: n_embd_v_gqa = 1024
llm_load_print_meta: f_norm_eps = 0.0e+00
llm_load_print_meta: f_norm_rms_eps = 1.0e-05
llm_load_print_meta: f_clamp_kqv = 0.0e+00
llm_load_print_meta: f_max_alibi_bias = 0.0e+00
llm_load_print_meta: f_logit_scale = 0.0e+00
llm_load_print_meta: n_ff = 13824
llm_load_print_meta: n_expert = 0
llm_load_print_meta: n_expert_used = 0
llm_load_print_meta: causal attn = 1
llm_load_print_meta: pooling type = 0
llm_load_print_meta: rope type = 2
llm_load_print_meta: rope scaling = linear
llm_load_print_meta: freq_base_train = 1000000.0
llm_load_print_meta: freq_scale_train = 1
llm_load_print_meta: n_ctx_orig_yarn = 131072
llm_load_print_meta: rope_finetuned = unknown
llm_load_print_meta: ssm_d_conv = 0
llm_load_print_meta: ssm_d_inner = 0
llm_load_print_meta: ssm_d_state = 0
llm_load_print_meta: ssm_dt_rank = 0
llm_load_print_meta: ssm_dt_b_c_rms = 0
llm_load_print_meta: model type = 14B
llm_load_print_meta: model ftype = F16
llm_load_print_meta: model params = 14.77 B
llm_load_print_meta: model size = 27.51 GiB (16.00 BPW)
llm_load_print_meta: general.name = Checkpoint 887 Merged
llm_load_print_meta: BOS token = 151646 '<｜begin▁of▁sentence｜>'
llm_load_print_meta: EOS token = 151643 '<｜end▁of▁sentence｜>'
llm_load_print_meta: EOT token = 151643 '<｜end▁of▁sentence｜>'
llm_load_print_meta: PAD token = 151643 '<｜end▁of▁sentence｜>'
llm_load_print_meta: LF token = 148848 'ÄĬ'
llm_load_print_meta: FIM PRE token = 151659 '<|fim_prefix|>'
llm_load_print_meta: FIM SUF token = 151661 '<|fim_suffix|>'
llm_load_print_meta: FIM MID token = 151660 '<|fim_middle|>'
llm_load_print_meta: FIM PAD token = 151662 '<|fim_pad|>'
llm_load_print_meta: FIM REP token = 151663 '<|repo_name|>'
llm_load_print_meta: FIM SEP token = 151664 '<|file_sep|>'
llm_load_print_meta: EOG token = 151643 '<｜end▁of▁sentence｜>'
llm_load_print_meta: EOG token = 151662 '<|fim_pad|>'
llm_load_print_meta: EOG token = 151663 '<|repo_name|>'
llm_load_print_meta: EOG token = 151664 '<|file_sep|>'
llm_load_print_meta: max token length = 256
llm_load_tensors: CPU model buffer size = 28173.21 MiB
load_all_data: no device found for buffer type CPU for async uploads
time=2025-02-21T23:18:32.271+08:00 level=INFO source=server.go:591 msg="waiting for server to become available" status="llm server loading model"
time=2025-02-21T23:18:33.022+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.05"
time=2025-02-21T23:18:33.523+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.11"
time=2025-02-21T23:18:33.773+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.13"
time=2025-02-21T23:18:34.023+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.15"
time=2025-02-21T23:18:34.274+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.17"
time=2025-02-21T23:18:34.524+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.18"
time=2025-02-21T23:18:34.774+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.20"
time=2025-02-21T23:18:35.024+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.22"
time=2025-02-21T23:18:35.275+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.24"
time=2025-02-21T23:18:35.525+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.26"
time=2025-02-21T23:18:35.775+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.28"
time=2025-02-21T23:18:36.025+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.30"
time=2025-02-21T23:18:36.275+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.32"
time=2025-02-21T23:18:36.526+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.34"
time=2025-02-21T23:18:36.777+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.36"
time=2025-02-21T23:18:37.028+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.38"
time=2025-02-21T23:18:37.278+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.40"
time=2025-02-21T23:18:37.528+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.42"
time=2025-02-21T23:18:37.778+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.44"
time=2025-02-21T23:18:38.028+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.45"
time=2025-02-21T23:18:38.279+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.47"
time=2025-02-21T23:18:38.530+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.50"
time=2025-02-21T23:18:38.780+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.52"
time=2025-02-21T23:18:39.030+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.54"
time=2025-02-21T23:18:39.280+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.56"
time=2025-02-21T23:18:39.531+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.58"
time=2025-02-21T23:18:39.781+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.59"
time=2025-02-21T23:18:40.032+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.61"
time=2025-02-21T23:18:40.282+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.63"
time=2025-02-21T23:18:40.532+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.65"
time=2025-02-21T23:18:40.783+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.67"
time=2025-02-21T23:18:41.033+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.69"
time=2025-02-21T23:18:41.283+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.71"
time=2025-02-21T23:18:41.534+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.72"
time=2025-02-21T23:18:41.784+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.75"
time=2025-02-21T23:18:42.035+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.77"
time=2025-02-21T23:18:42.285+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.79"
time=2025-02-21T23:18:42.535+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.80"
time=2025-02-21T23:18:42.786+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.82"
time=2025-02-21T23:18:43.036+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.85"
time=2025-02-21T23:18:43.286+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.86"
time=2025-02-21T23:18:43.536+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.88"
time=2025-02-21T23:18:43.787+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.91"
time=2025-02-21T23:18:44.037+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.93"
time=2025-02-21T23:18:44.288+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.95"
time=2025-02-21T23:18:44.538+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.97"
time=2025-02-21T23:18:44.788+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.99"
llama_new_context_with_model: n_seq_max = 1
llama_new_context_with_model: n_ctx = 2048
llama_new_context_with_model: n_ctx_per_seq = 2048
llama_new_context_with_model: n_batch = 512
llama_new_context_with_model: n_ubatch = 512
llama_new_context_with_model: flash_attn = 0
llama_new_context_with_model: freq_base = 1000000.0
llama_new_context_with_model: freq_scale = 1
llama_new_context_with_model: n_ctx_per_seq (2048) < n_ctx_train (131072) -- the full capacity of the model will not be utilized
llama_kv_cache_init: kv_size = 2048, offload = 1, type_k = 'f16', type_v = 'f16', n_layer = 48, can_shift = 1
llama_kv_cache_init: layer 0: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 1: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 2: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 3: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 4: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 5: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 6: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 7: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 8: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 9: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 10: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 11: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 12: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 13: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 14: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 15: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 16: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 17: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 18: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 19: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 20: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 21: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 22: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 23: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 24: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 25: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 26: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 27: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 28: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 29: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 30: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 31: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 32: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 33: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 34: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 35: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 36: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 37: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 38: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 39: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 40: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 41: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 42: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 43: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 44: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 45: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 46: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: layer 47: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024
llama_kv_cache_init: CPU KV buffer size = 384.00 MiB
llama_new_context_with_model: KV self size = 384.00 MiB, K (f16): 192.00 MiB, V (f16): 192.00 MiB
llama_new_context_with_model: CPU output buffer size = 0.60 MiB
llama_new_context_with_model: CPU compute buffer size = 307.00 MiB
llama_new_context_with_model: graph nodes = 1686
llama_new_context_with_model: graph splits = 1
time=2025-02-21T23:18:45.039+08:00 level=INFO source=server.go:596 msg="llama runner started in 13.02 seconds"
time=2025-02-21T23:18:45.039+08:00 level=DEBUG source=sched.go:463 msg="finished setting up runner" model=D:\ollama\models\blobs\sha256-553aa261cfb6856c595c9fefdb5453b98fdef331bf2ca918a5e0a23aa254d022
[GIN] 2025/02/21 - 23:18:45 | 200 | 13.1287157s | 127.0.0.1 | POST "/api/generate"
time=2025-02-21T23:18:45.039+08:00 level=DEBUG source=sched.go:467 msg="context for request finished"
time=2025-02-21T23:18:45.039+08:00 level=DEBUG source=sched.go:340 msg="runner with non-zero duration has gone idle, adding timer" modelPath=D:\ollama\models\blobs\sha256-553aa261cfb6856c595c9fefdb5453b98fdef331bf2ca918a5e0a23aa254d022 duration=5m0s
time=2025-02-21T23:18:45.039+08:00 level=DEBUG source=sched.go:358 msg="after processing request finished event" modelPath=D:\ollama\models\blobs\sha256-553aa261cfb6856c595c9fefdb5453b98fdef331bf2ca918a5e0a23aa254d022 refCount=0
time=2025-02-21T23:18:54.108+08:00 level=DEBUG source=sched.go:576 msg="evaluating already loaded" model=D:\ollama\models\blobs\sha256-553aa261cfb6856c595c9fefdb5453b98fdef331bf2ca918a5e0a23aa254d022
time=2025-02-21T23:18:54.108+08:00 level=DEBUG source=routes.go:1462 msg="chat request" images=0 prompt=hi
time=2025-02-21T23:18:54.108+08:00 level=DEBUG source=cache.go:104 msg="loading cache slot" id=0 cache=0 prompt=2 used=0 remaining=2

<!-- gh-comment-id:2674844067 --> @Hsq12138 commented on GitHub (Feb 21, 2025): > `OLLAMA_GPU_LAYERS` is not an ollama environment variable. > > ``` > 2月 21 13:33:11 tc ollama[285585]: time=2025-02-21T13:33:11.387+08:00 level=DEBUG source=ggml.go:89 msg="ggml backend load all from path" path=/usr/local/bin > ``` > > ollama couldn't find an backends to load. What's the output of: > > ``` > ls -l /usr/local/lib /usr/local/lib/ollama > ``` > > How did you install ollama? 2025/02/21 23:08:26 routes.go:1187: INFO server config env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_DEBUG:true OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 OLLAMA_INTEL_GPU:false 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:D:\\ollama\\models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:0 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://*] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES:]" time=2025-02-21T23:08:26.801+08:00 level=INFO source=images.go:432 msg="total blobs: 2" time=2025-02-21T23:08:26.802+08:00 level=INFO source=images.go:439 msg="total unused blobs removed: 0" time=2025-02-21T23:08:26.802+08:00 level=INFO source=routes.go:1238 msg="Listening on 127.0.0.1:11434 (version 0.5.12-rc1)" time=2025-02-21T23:08:26.802+08:00 level=DEBUG source=sched.go:106 msg="starting llm scheduler" time=2025-02-21T23:08:26.803+08:00 level=INFO source=gpu.go:217 msg="looking for compatible GPUs" time=2025-02-21T23:08:26.803+08:00 level=INFO source=gpu_windows.go:167 msg=packages count=1 time=2025-02-21T23:08:26.803+08:00 level=INFO source=gpu_windows.go:214 msg="" package=0 cores=6 efficiency=0 threads=12 time=2025-02-21T23:08:26.803+08:00 level=DEBUG source=gpu.go:98 msg="searching for GPU discovery libraries for NVIDIA" time=2025-02-21T23:08:26.803+08:00 level=DEBUG source=gpu.go:501 msg="Searching for GPU library" name=nvml.dll time=2025-02-21T23:08:26.803+08:00 level=DEBUG source=gpu.go:525 msg="gpu library search" globs="[C:\\Users\\zrway\\AppData\\Local\\Programs\\Ollama\\lib\\ollama\\nvml.dll C:\\Program Files\\NVIDIA\\CUDNN\\v9.7\\bin\\nvml.dll C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v12.8\\bin\\nvml.dll C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v12.8\\libnvvp\\nvml.dll C:\\Users\\zrway\\AppData\\Local\\Programs\\Ollama\\nvml.dll C:\\Program Files\\Common Files\\Oracle\\Java\\javapath\\nvml.dll C:\\Windows\\system32\\nvml.dll C:\\Windows\\nvml.dll C:\\Windows\\System32\\Wbem\\nvml.dll C:\\Windows\\System32\\WindowsPowerShell\\v1.0\\nvml.dll C:\\Windows\\System32\\OpenSSH\\nvml.dll C:\\Program Files (x86)\\NVIDIA Corporation\\PhysX\\Common\\nvml.dll C:\\Program Files\\Bandizip\\nvml.dll C:\\Program Files\\dotnet\\nvml.dll C:\\Program Files\\Git\\cmd\\nvml.dll D:\\ai\\ffmpeg-2024-03-28-git-5d71f97e0e-full_build\\bin\\nvml.dll C:\\Program Files\\NVIDIA Corporation\\NVIDIA app\\NvDLISR\\nvml.dll C:\\Users\\zrway\\AppData\\Local\\Microsoft\\WindowsApps\\python.exe\\nvml.dll C:\\Users\\zrway\\AppData\\Local\\Packages\\PythonSoftwareFoundation.Python.3.11_qbz5n2kfra8p0\\LocalCache\\local-packages\\Python311\\Scripts\\nvml.dll C:\\Program Files\\NVIDIA Corporation\\Nsight Compute 2025.1.0\\nvml.dll C:\\Program Files\\MySQL\\MySQL Shell 8.0\\bin\\nvml.dll C:\\Users\\zrway\\AppData\\Local\\Microsoft\\WindowsApps\\nvml.dll C:\\Users\\zrway\\AppData\\Local\\Programs\\Microsoft VS Code\\bin\\nvml.dll D:\\ai\\ffmpeg-2024-03-28-git-5d71f97e0e-essentials_build\\bin\\nvml.dll C:\\Users\\zrway\\AppData\\Local\\Programs\\Ollama\\nvml.dll C:\\Users\\zrway\\AppData\\Local\\Programs\\Ollama\\nvml.dll C:\\Users\\zrway\\.lmstudio\\bin\\nvml.dll c:\\Windows\\System32\\nvml.dll]" time=2025-02-21T23:08:26.804+08:00 level=DEBUG source=gpu.go:529 msg="skipping PhysX cuda library path" path="C:\\Program Files (x86)\\NVIDIA Corporation\\PhysX\\Common\\nvml.dll" time=2025-02-21T23:08:26.805+08:00 level=DEBUG source=gpu.go:558 msg="discovered GPU libraries" paths="[C:\\Windows\\system32\\nvml.dll c:\\Windows\\System32\\nvml.dll]" time=2025-02-21T23:08:26.839+08:00 level=DEBUG source=gpu.go:111 msg="nvidia-ml loaded" library=C:\Windows\system32\nvml.dll time=2025-02-21T23:08:26.839+08:00 level=DEBUG source=gpu.go:501 msg="Searching for GPU library" name=nvcuda.dll time=2025-02-21T23:08:26.839+08:00 level=DEBUG source=gpu.go:525 msg="gpu library search" globs="[C:\\Users\\zrway\\AppData\\Local\\Programs\\Ollama\\lib\\ollama\\nvcuda.dll C:\\Program Files\\NVIDIA\\CUDNN\\v9.7\\bin\\nvcuda.dll C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v12.8\\bin\\nvcuda.dll C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v12.8\\libnvvp\\nvcuda.dll C:\\Users\\zrway\\AppData\\Local\\Programs\\Ollama\\nvcuda.dll C:\\Program Files\\Common Files\\Oracle\\Java\\javapath\\nvcuda.dll C:\\Windows\\system32\\nvcuda.dll C:\\Windows\\nvcuda.dll C:\\Windows\\System32\\Wbem\\nvcuda.dll C:\\Windows\\System32\\WindowsPowerShell\\v1.0\\nvcuda.dll C:\\Windows\\System32\\OpenSSH\\nvcuda.dll C:\\Program Files (x86)\\NVIDIA Corporation\\PhysX\\Common\\nvcuda.dll C:\\Program Files\\Bandizip\\nvcuda.dll C:\\Program Files\\dotnet\\nvcuda.dll C:\\Program Files\\Git\\cmd\\nvcuda.dll D:\\ai\\ffmpeg-2024-03-28-git-5d71f97e0e-full_build\\bin\\nvcuda.dll C:\\Program Files\\NVIDIA Corporation\\NVIDIA app\\NvDLISR\\nvcuda.dll C:\\Users\\zrway\\AppData\\Local\\Microsoft\\WindowsApps\\python.exe\\nvcuda.dll C:\\Users\\zrway\\AppData\\Local\\Packages\\PythonSoftwareFoundation.Python.3.11_qbz5n2kfra8p0\\LocalCache\\local-packages\\Python311\\Scripts\\nvcuda.dll C:\\Program Files\\NVIDIA Corporation\\Nsight Compute 2025.1.0\\nvcuda.dll C:\\Program Files\\MySQL\\MySQL Shell 8.0\\bin\\nvcuda.dll C:\\Users\\zrway\\AppData\\Local\\Microsoft\\WindowsApps\\nvcuda.dll C:\\Users\\zrway\\AppData\\Local\\Programs\\Microsoft VS Code\\bin\\nvcuda.dll D:\\ai\\ffmpeg-2024-03-28-git-5d71f97e0e-essentials_build\\bin\\nvcuda.dll C:\\Users\\zrway\\AppData\\Local\\Programs\\Ollama\\nvcuda.dll C:\\Users\\zrway\\AppData\\Local\\Programs\\Ollama\\nvcuda.dll C:\\Users\\zrway\\.lmstudio\\bin\\nvcuda.dll c:\\windows\\system*\\nvcuda.dll]" time=2025-02-21T23:08:26.840+08:00 level=DEBUG source=gpu.go:529 msg="skipping PhysX cuda library path" path="C:\\Program Files (x86)\\NVIDIA Corporation\\PhysX\\Common\\nvcuda.dll" time=2025-02-21T23:08:26.841+08:00 level=DEBUG source=gpu.go:558 msg="discovered GPU libraries" paths=[C:\Windows\system32\nvcuda.dll] initializing C:\Windows\system32\nvcuda.dll dlsym: cuInit - 00007FF97B775F80 dlsym: cuDriverGetVersion - 00007FF97B776020 dlsym: cuDeviceGetCount - 00007FF97B776816 dlsym: cuDeviceGet - 00007FF97B776810 dlsym: cuDeviceGetAttribute - 00007FF97B776170 dlsym: cuDeviceGetUuid - 00007FF97B776822 dlsym: cuDeviceGetName - 00007FF97B77681C dlsym: cuCtxCreate_v3 - 00007FF97B776894 dlsym: cuMemGetInfo_v2 - 00007FF97B776996 dlsym: cuCtxDestroy - 00007FF97B7768A6 calling cuInit calling cuDriverGetVersion raw version 0x2f30 CUDA driver version: 12.8 calling cuDeviceGetCount device count 1 time=2025-02-21T23:08:26.893+08:00 level=DEBUG source=gpu.go:125 msg="detected GPUs" count=1 library=C:\Windows\system32\nvcuda.dll [GPU-ecc0382b-7d7c-7b61-8572-a21b10ac9fcd] CUDA totalMem 16375 mb [GPU-ecc0382b-7d7c-7b61-8572-a21b10ac9fcd] CUDA freeMem 15035 mb [GPU-ecc0382b-7d7c-7b61-8572-a21b10ac9fcd] Compute Capability 8.9 time=2025-02-21T23:08:26.975+08:00 level=INFO source=gpu.go:319 msg="detected OS VRAM overhead" id=GPU-ecc0382b-7d7c-7b61-8572-a21b10ac9fcd library=cuda compute=8.9 driver=12.8 name="NVIDIA GeForce RTX 4080 SUPER" overhead="265.6 MiB" time=2025-02-21T23:08:26.976+08:00 level=DEBUG source=amd_windows.go:34 msg="unable to load amdhip64_6.dll, please make sure to upgrade to the latest amd driver: The file cannot be accessed by the system." releasing cuda driver library releasing nvml library time=2025-02-21T23:08:26.977+08:00 level=INFO source=types.go:130 msg="inference compute" id=GPU-ecc0382b-7d7c-7b61-8572-a21b10ac9fcd library=cuda variant=v12 compute=8.9 driver=12.8 name="NVIDIA GeForce RTX 4080 SUPER" total="16.0 GiB" available="14.7 GiB" [GIN] 2025/02/21 - 23:18:16 | 200 | 0s | 127.0.0.1 | HEAD "/" [GIN] 2025/02/21 - 23:18:16 | 200 | 1.0301ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/02/21 - 23:18:31 | 200 | 0s | 127.0.0.1 | HEAD "/" [GIN] 2025/02/21 - 23:18:31 | 200 | 10.8992ms | 127.0.0.1 | POST "/api/show" time=2025-02-21T23:18:31.919+08:00 level=DEBUG source=gpu.go:391 msg="updating system memory data" before.total="95.8 GiB" before.free="87.3 GiB" before.free_swap="89.6 GiB" now.total="95.8 GiB" now.free="86.2 GiB" now.free_swap="87.3 GiB" time=2025-02-21T23:18:31.928+08:00 level=DEBUG source=gpu.go:441 msg="updating cuda memory data" gpu=GPU-ecc0382b-7d7c-7b61-8572-a21b10ac9fcd name="NVIDIA GeForce RTX 4080 SUPER" overhead="265.6 MiB" before.total="16.0 GiB" before.free="14.7 GiB" now.total="16.0 GiB" now.free="14.4 GiB" now.used="1.3 GiB" releasing nvml library time=2025-02-21T23:18:31.928+08:00 level=DEBUG source=sched.go:182 msg="updating default concurrency" OLLAMA_MAX_LOADED_MODELS=3 gpu_count=1 time=2025-02-21T23:18:31.946+08:00 level=DEBUG source=sched.go:225 msg="loading first model" model=D:\ollama\models\blobs\sha256-553aa261cfb6856c595c9fefdb5453b98fdef331bf2ca918a5e0a23aa254d022 time=2025-02-21T23:18:31.946+08:00 level=DEBUG source=memory.go:108 msg=evaluating library=cuda gpu_count=1 available="[14.4 GiB]" time=2025-02-21T23:18:31.946+08:00 level=WARN source=ggml.go:132 msg="key not found" key=qwen2.attention.key_length default=128 time=2025-02-21T23:18:31.946+08:00 level=WARN source=ggml.go:132 msg="key not found" key=qwen2.attention.value_length default=128 time=2025-02-21T23:18:31.947+08:00 level=DEBUG source=memory.go:108 msg=evaluating library=cuda gpu_count=1 available="[14.4 GiB]" time=2025-02-21T23:18:31.947+08:00 level=WARN source=ggml.go:132 msg="key not found" key=qwen2.attention.key_length default=128 time=2025-02-21T23:18:31.947+08:00 level=WARN source=ggml.go:132 msg="key not found" key=qwen2.attention.value_length default=128 time=2025-02-21T23:18:31.948+08:00 level=DEBUG source=memory.go:108 msg=evaluating library=cuda gpu_count=1 available="[14.4 GiB]" time=2025-02-21T23:18:31.948+08:00 level=WARN source=ggml.go:132 msg="key not found" key=qwen2.attention.key_length default=128 time=2025-02-21T23:18:31.948+08:00 level=WARN source=ggml.go:132 msg="key not found" key=qwen2.attention.value_length default=128 time=2025-02-21T23:18:31.948+08:00 level=DEBUG source=memory.go:108 msg=evaluating library=cuda gpu_count=1 available="[14.4 GiB]" time=2025-02-21T23:18:31.948+08:00 level=WARN source=ggml.go:132 msg="key not found" key=qwen2.attention.key_length default=128 time=2025-02-21T23:18:31.948+08:00 level=WARN source=ggml.go:132 msg="key not found" key=qwen2.attention.value_length default=128 time=2025-02-21T23:18:31.948+08:00 level=DEBUG source=gpu.go:391 msg="updating system memory data" before.total="95.8 GiB" before.free="86.2 GiB" before.free_swap="87.3 GiB" now.total="95.8 GiB" now.free="86.2 GiB" now.free_swap="87.3 GiB" time=2025-02-21T23:18:31.956+08:00 level=DEBUG source=gpu.go:441 msg="updating cuda memory data" gpu=GPU-ecc0382b-7d7c-7b61-8572-a21b10ac9fcd name="NVIDIA GeForce RTX 4080 SUPER" overhead="265.6 MiB" before.total="16.0 GiB" before.free="14.4 GiB" now.total="16.0 GiB" now.free="14.4 GiB" now.used="1.3 GiB" releasing nvml library time=2025-02-21T23:18:31.956+08:00 level=INFO source=server.go:97 msg="system memory" total="95.8 GiB" free="86.2 GiB" free_swap="87.3 GiB" time=2025-02-21T23:18:31.956+08:00 level=DEBUG source=memory.go:108 msg=evaluating library=cuda gpu_count=1 available="[14.4 GiB]" time=2025-02-21T23:18:31.956+08:00 level=WARN source=ggml.go:132 msg="key not found" key=qwen2.attention.key_length default=128 time=2025-02-21T23:18:31.956+08:00 level=WARN source=ggml.go:132 msg="key not found" key=qwen2.attention.value_length default=128 time=2025-02-21T23:18:31.957+08:00 level=INFO source=server.go:130 msg=offload library=cuda layers.requested=-1 layers.model=49 layers.offload=24 layers.split="" memory.available="[14.4 GiB]" memory.gpu_overhead="0 B" memory.required.full="28.3 GiB" memory.required.partial="14.4 GiB" memory.required.kv="384.0 MiB" memory.required.allocations="[14.4 GiB]" memory.weights.total="25.0 GiB" memory.weights.repeating="23.5 GiB" memory.weights.nonrepeating="1.5 GiB" memory.graph.full="307.0 MiB" memory.graph.partial="916.1 MiB" time=2025-02-21T23:18:31.957+08:00 level=DEBUG source=server.go:259 msg="compatible gpu libraries" compatible="[cuda_v12 cuda_v11]" time=2025-02-21T23:18:31.963+08:00 level=DEBUG source=server.go:302 msg="adding gpu library" path=C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\cuda_v12 time=2025-02-21T23:18:31.963+08:00 level=DEBUG source=server.go:310 msg="adding gpu dependency paths" paths=[C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\cuda_v12] time=2025-02-21T23:18:31.963+08:00 level=INFO source=server.go:380 msg="starting llama server" cmd="C:\\Users\\zrway\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --model D:\\ollama\\models\\blobs\\sha256-553aa261cfb6856c595c9fefdb5453b98fdef331bf2ca918a5e0a23aa254d022 --ctx-size 2048 --batch-size 512 --n-gpu-layers 24 --verbose --threads 6 --no-mmap --parallel 1 --port 50841" time=2025-02-21T23:18:31.963+08:00 level=DEBUG source=server.go:398 msg=subprocess environment="[CUDA_PATH=C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v12.8 CUDA_PATH_V12_8=C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v12.8 PATH=C:\\Users\\zrway\\AppData\\Local\\Programs\\Ollama\\lib\\ollama\\cuda_v12;C:\\Program Files\\NVIDIA\\CUDNN\\v9.7\\bin;C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v12.8\\bin;C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v12.8\\libnvvp;;C:\\Program Files\\Common Files\\Oracle\\Java\\javapath;C:\\Windows\\system32;C:\\Windows;C:\\Windows\\System32\\Wbem;C:\\Windows\\System32\\WindowsPowerShell\\v1.0\\;C:\\Windows\\System32\\OpenSSH\\;C:\\Program Files (x86)\\NVIDIA Corporation\\PhysX\\Common;C:\\Program Files\\Bandizip\\;C:\\Program Files\\dotnet\\;C:\\Program Files\\Git\\cmd;D:\\ai\\ffmpeg-2024-03-28-git-5d71f97e0e-full_build\\bin;C:\\Program Files\\NVIDIA Corporation\\NVIDIA app\\NvDLISR;C:\\Users\\zrway\\AppData\\Local\\Microsoft\\WindowsApps\\python.exe;C:\\Users\\zrway\\AppData\\Local\\Packages\\PythonSoftwareFoundation.Python.3.11_qbz5n2kfra8p0\\LocalCache\\local-packages\\Python311\\Scripts;C:\\Program Files\\NVIDIA Corporation\\Nsight Compute 2025.1.0\\;C:\\Program Files\\MySQL\\MySQL Shell 8.0\\bin\\;C:\\Users\\zrway\\AppData\\Local\\Microsoft\\WindowsApps;C:\\Users\\zrway\\AppData\\Local\\Programs\\Microsoft VS Code\\bin;D:\\ai\\ffmpeg-2024-03-28-git-5d71f97e0e-essentials_build\\bin;;C:\\Users\\zrway\\AppData\\Local\\Programs\\Ollama;C:\\Users\\zrway\\.lmstudio\\bin;C:\\Users\\zrway\\AppData\\Local\\Programs\\Ollama\\lib\\ollama\\cuda_v12;C:\\Users\\zrway\\AppData\\Local\\Programs\\Ollama\\lib\\ollama CUDA_VISIBLE_DEVICES=GPU-ecc0382b-7d7c-7b61-8572-a21b10ac9fcd]" time=2025-02-21T23:18:32.020+08:00 level=INFO source=sched.go:450 msg="loaded runners" count=1 time=2025-02-21T23:18:32.020+08:00 level=INFO source=server.go:557 msg="waiting for llama runner to start responding" time=2025-02-21T23:18:32.020+08:00 level=INFO source=server.go:591 msg="waiting for server to become available" status="llm server error" time=2025-02-21T23:18:32.035+08:00 level=INFO source=runner.go:932 msg="starting go runner" time=2025-02-21T23:18:32.041+08:00 level=DEBUG source=ggml.go:89 msg="ggml backend load all from path" path=C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\cuda_v12 time=2025-02-21T23:18:32.051+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path="C:\\Program Files\\NVIDIA\\CUDNN\\v9.7\\bin" time=2025-02-21T23:18:32.051+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path="C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v12.8\\bin" time=2025-02-21T23:18:32.051+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path="C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v12.8\\libnvvp" time=2025-02-21T23:18:32.051+08:00 level=DEBUG source=ggml.go:89 msg="ggml backend load all from path" path=C:\Users\zrway\AppData\Local\Programs\Ollama time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path="C:\\Program Files\\Common Files\\Oracle\\Java\\javapath" time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path=C:\Windows\system32 time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path=C:\Windows time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path=C:\Windows\System32\Wbem time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path=C:\Windows\System32\WindowsPowerShell\v1.0 time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path=C:\Windows\System32\OpenSSH time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path="C:\\Program Files (x86)\\NVIDIA Corporation\\PhysX\\Common" time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path="C:\\Program Files\\Bandizip" time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path="C:\\Program Files\\dotnet" time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path="C:\\Program Files\\Git\\cmd" time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path=D:\ai\ffmpeg-2024-03-28-git-5d71f97e0e-full_build\bin time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path="C:\\Program Files\\NVIDIA Corporation\\NVIDIA app\\NvDLISR" time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path=C:\Users\zrway\AppData\Local\Microsoft\WindowsApps\python.exe time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path=C:\Users\zrway\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.11_qbz5n2kfra8p0\LocalCache\local-packages\Python311\Scripts time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path="C:\\Program Files\\NVIDIA Corporation\\Nsight Compute 2025.1.0" time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path="C:\\Program Files\\MySQL\\MySQL Shell 8.0\\bin" time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path=C:\Users\zrway\AppData\Local\Microsoft\WindowsApps time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path="C:\\Users\\zrway\\AppData\\Local\\Programs\\Microsoft VS Code\\bin" time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path=D:\ai\ffmpeg-2024-03-28-git-5d71f97e0e-essentials_build\bin time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:83 msg="skipping path which is not part of ollama" path=C:\Users\zrway\.lmstudio\bin time=2025-02-21T23:18:32.052+08:00 level=DEBUG source=ggml.go:89 msg="ggml backend load all from path" path=C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-alderlake.dll ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-icelake.dll ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-sandybridge.dll ggml_backend_load_best: failed to load C:\Users\zrway\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-skylakex.dll time=2025-02-21T23:18:32.060+08:00 level=INFO source=runner.go:935 msg=system info="CPU : LLAMAFILE = 1 | CPU : LLAMAFILE = 1 | cgo(clang)" threads=6 time=2025-02-21T23:18:32.074+08:00 level=INFO source=runner.go:993 msg="Server listening on 127.0.0.1:50841" llama_model_loader: loaded meta data with 25 key-value pairs and 579 tensors from D:\ollama\models\blobs\sha256-553aa261cfb6856c595c9fefdb5453b98fdef331bf2ca918a5e0a23aa254d022 (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 = Checkpoint 887 Merged llama_model_loader: - kv 3: general.size_label str = 15B llama_model_loader: - kv 4: qwen2.block_count u32 = 48 llama_model_loader: - kv 5: qwen2.context_length u32 = 131072 llama_model_loader: - kv 6: qwen2.embedding_length u32 = 5120 llama_model_loader: - kv 7: qwen2.feed_forward_length u32 = 13824 llama_model_loader: - kv 8: qwen2.attention.head_count u32 = 40 llama_model_loader: - kv 9: qwen2.attention.head_count_kv u32 = 8 llama_model_loader: - kv 10: qwen2.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 11: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 12: general.file_type u32 = 1 llama_model_loader: - kv 13: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 14: tokenizer.ggml.pre str = deepseek-r1-qwen llama_model_loader: - kv 15: tokenizer.ggml.tokens arr[str,152064] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 16: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 17: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 18: tokenizer.ggml.bos_token_id u32 = 151646 llama_model_loader: - kv 19: tokenizer.ggml.eos_token_id u32 = 151643 llama_model_loader: - kv 20: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 21: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 22: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 23: tokenizer.chat_template str = {% if not add_generation_prompt is de... llama_model_loader: - kv 24: general.quantization_version u32 = 2 llama_model_loader: - type f32: 241 tensors llama_model_loader: - type f16: 338 tensors llm_load_vocab: missing or unrecognized pre-tokenizer type, using: 'default' llm_load_vocab: control token: 151659 '<|fim_prefix|>' is not marked as EOG llm_load_vocab: control token: 151656 '<|video_pad|>' is not marked as EOG llm_load_vocab: control token: 151655 '<|image_pad|>' is not marked as EOG llm_load_vocab: control token: 151653 '<|vision_end|>' is not marked as EOG llm_load_vocab: control token: 151652 '<|vision_start|>' is not marked as EOG llm_load_vocab: control token: 151651 '<|quad_end|>' is not marked as EOG llm_load_vocab: control token: 151646 '<｜begin▁of▁sentence｜>' is not marked as EOG llm_load_vocab: control token: 151644 '<｜User｜>' is not marked as EOG llm_load_vocab: control token: 151661 '<|fim_suffix|>' is not marked as EOG llm_load_vocab: control token: 151660 '<|fim_middle|>' is not marked as EOG llm_load_vocab: control token: 151654 '<|vision_pad|>' is not marked as EOG llm_load_vocab: control token: 151650 '<|quad_start|>' is not marked as EOG llm_load_vocab: control token: 151647 '<|EOT|>' is not marked as EOG llm_load_vocab: control token: 151643 '<｜end▁of▁sentence｜>' is not marked as EOG llm_load_vocab: control token: 151645 '<｜Assistant｜>' is not marked as EOG llm_load_vocab: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect llm_load_vocab: special tokens cache size = 22 llm_load_vocab: token to piece cache size = 0.9310 MB llm_load_print_meta: format = GGUF V3 (latest) llm_load_print_meta: arch = qwen2 llm_load_print_meta: vocab type = BPE llm_load_print_meta: n_vocab = 152064 llm_load_print_meta: n_merges = 151387 llm_load_print_meta: vocab_only = 0 llm_load_print_meta: n_ctx_train = 131072 llm_load_print_meta: n_embd = 5120 llm_load_print_meta: n_layer = 48 llm_load_print_meta: n_head = 40 llm_load_print_meta: n_head_kv = 8 llm_load_print_meta: n_rot = 128 llm_load_print_meta: n_swa = 0 llm_load_print_meta: n_embd_head_k = 128 llm_load_print_meta: n_embd_head_v = 128 llm_load_print_meta: n_gqa = 5 llm_load_print_meta: n_embd_k_gqa = 1024 llm_load_print_meta: n_embd_v_gqa = 1024 llm_load_print_meta: f_norm_eps = 0.0e+00 llm_load_print_meta: f_norm_rms_eps = 1.0e-05 llm_load_print_meta: f_clamp_kqv = 0.0e+00 llm_load_print_meta: f_max_alibi_bias = 0.0e+00 llm_load_print_meta: f_logit_scale = 0.0e+00 llm_load_print_meta: n_ff = 13824 llm_load_print_meta: n_expert = 0 llm_load_print_meta: n_expert_used = 0 llm_load_print_meta: causal attn = 1 llm_load_print_meta: pooling type = 0 llm_load_print_meta: rope type = 2 llm_load_print_meta: rope scaling = linear llm_load_print_meta: freq_base_train = 1000000.0 llm_load_print_meta: freq_scale_train = 1 llm_load_print_meta: n_ctx_orig_yarn = 131072 llm_load_print_meta: rope_finetuned = unknown llm_load_print_meta: ssm_d_conv = 0 llm_load_print_meta: ssm_d_inner = 0 llm_load_print_meta: ssm_d_state = 0 llm_load_print_meta: ssm_dt_rank = 0 llm_load_print_meta: ssm_dt_b_c_rms = 0 llm_load_print_meta: model type = 14B llm_load_print_meta: model ftype = F16 llm_load_print_meta: model params = 14.77 B llm_load_print_meta: model size = 27.51 GiB (16.00 BPW) llm_load_print_meta: general.name = Checkpoint 887 Merged llm_load_print_meta: BOS token = 151646 '<｜begin▁of▁sentence｜>' llm_load_print_meta: EOS token = 151643 '<｜end▁of▁sentence｜>' llm_load_print_meta: EOT token = 151643 '<｜end▁of▁sentence｜>' llm_load_print_meta: PAD token = 151643 '<｜end▁of▁sentence｜>' llm_load_print_meta: LF token = 148848 'ÄĬ' llm_load_print_meta: FIM PRE token = 151659 '<|fim_prefix|>' llm_load_print_meta: FIM SUF token = 151661 '<|fim_suffix|>' llm_load_print_meta: FIM MID token = 151660 '<|fim_middle|>' llm_load_print_meta: FIM PAD token = 151662 '<|fim_pad|>' llm_load_print_meta: FIM REP token = 151663 '<|repo_name|>' llm_load_print_meta: FIM SEP token = 151664 '<|file_sep|>' llm_load_print_meta: EOG token = 151643 '<｜end▁of▁sentence｜>' llm_load_print_meta: EOG token = 151662 '<|fim_pad|>' llm_load_print_meta: EOG token = 151663 '<|repo_name|>' llm_load_print_meta: EOG token = 151664 '<|file_sep|>' llm_load_print_meta: max token length = 256 llm_load_tensors: CPU model buffer size = 28173.21 MiB load_all_data: no device found for buffer type CPU for async uploads time=2025-02-21T23:18:32.271+08:00 level=INFO source=server.go:591 msg="waiting for server to become available" status="llm server loading model" time=2025-02-21T23:18:33.022+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.05" time=2025-02-21T23:18:33.523+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.11" time=2025-02-21T23:18:33.773+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.13" time=2025-02-21T23:18:34.023+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.15" time=2025-02-21T23:18:34.274+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.17" time=2025-02-21T23:18:34.524+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.18" time=2025-02-21T23:18:34.774+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.20" time=2025-02-21T23:18:35.024+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.22" time=2025-02-21T23:18:35.275+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.24" time=2025-02-21T23:18:35.525+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.26" time=2025-02-21T23:18:35.775+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.28" time=2025-02-21T23:18:36.025+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.30" time=2025-02-21T23:18:36.275+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.32" time=2025-02-21T23:18:36.526+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.34" time=2025-02-21T23:18:36.777+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.36" time=2025-02-21T23:18:37.028+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.38" time=2025-02-21T23:18:37.278+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.40" time=2025-02-21T23:18:37.528+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.42" time=2025-02-21T23:18:37.778+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.44" time=2025-02-21T23:18:38.028+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.45" time=2025-02-21T23:18:38.279+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.47" time=2025-02-21T23:18:38.530+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.50" time=2025-02-21T23:18:38.780+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.52" time=2025-02-21T23:18:39.030+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.54" time=2025-02-21T23:18:39.280+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.56" time=2025-02-21T23:18:39.531+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.58" time=2025-02-21T23:18:39.781+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.59" time=2025-02-21T23:18:40.032+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.61" time=2025-02-21T23:18:40.282+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.63" time=2025-02-21T23:18:40.532+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.65" time=2025-02-21T23:18:40.783+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.67" time=2025-02-21T23:18:41.033+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.69" time=2025-02-21T23:18:41.283+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.71" time=2025-02-21T23:18:41.534+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.72" time=2025-02-21T23:18:41.784+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.75" time=2025-02-21T23:18:42.035+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.77" time=2025-02-21T23:18:42.285+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.79" time=2025-02-21T23:18:42.535+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.80" time=2025-02-21T23:18:42.786+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.82" time=2025-02-21T23:18:43.036+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.85" time=2025-02-21T23:18:43.286+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.86" time=2025-02-21T23:18:43.536+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.88" time=2025-02-21T23:18:43.787+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.91" time=2025-02-21T23:18:44.037+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.93" time=2025-02-21T23:18:44.288+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.95" time=2025-02-21T23:18:44.538+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.97" time=2025-02-21T23:18:44.788+08:00 level=DEBUG source=server.go:602 msg="model load progress 0.99" llama_new_context_with_model: n_seq_max = 1 llama_new_context_with_model: n_ctx = 2048 llama_new_context_with_model: n_ctx_per_seq = 2048 llama_new_context_with_model: n_batch = 512 llama_new_context_with_model: n_ubatch = 512 llama_new_context_with_model: flash_attn = 0 llama_new_context_with_model: freq_base = 1000000.0 llama_new_context_with_model: freq_scale = 1 llama_new_context_with_model: n_ctx_per_seq (2048) < n_ctx_train (131072) -- the full capacity of the model will not be utilized llama_kv_cache_init: kv_size = 2048, offload = 1, type_k = 'f16', type_v = 'f16', n_layer = 48, can_shift = 1 llama_kv_cache_init: layer 0: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 1: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 2: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 3: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 4: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 5: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 6: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 7: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 8: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 9: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 10: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 11: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 12: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 13: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 14: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 15: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 16: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 17: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 18: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 19: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 20: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 21: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 22: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 23: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 24: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 25: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 26: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 27: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 28: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 29: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 30: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 31: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 32: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 33: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 34: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 35: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 36: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 37: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 38: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 39: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 40: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 41: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 42: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 43: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 44: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 45: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 46: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: layer 47: n_embd_k_gqa = 1024, n_embd_v_gqa = 1024 llama_kv_cache_init: CPU KV buffer size = 384.00 MiB llama_new_context_with_model: KV self size = 384.00 MiB, K (f16): 192.00 MiB, V (f16): 192.00 MiB llama_new_context_with_model: CPU output buffer size = 0.60 MiB llama_new_context_with_model: CPU compute buffer size = 307.00 MiB llama_new_context_with_model: graph nodes = 1686 llama_new_context_with_model: graph splits = 1 time=2025-02-21T23:18:45.039+08:00 level=INFO source=server.go:596 msg="llama runner started in 13.02 seconds" time=2025-02-21T23:18:45.039+08:00 level=DEBUG source=sched.go:463 msg="finished setting up runner" model=D:\ollama\models\blobs\sha256-553aa261cfb6856c595c9fefdb5453b98fdef331bf2ca918a5e0a23aa254d022 [GIN] 2025/02/21 - 23:18:45 | 200 | 13.1287157s | 127.0.0.1 | POST "/api/generate" time=2025-02-21T23:18:45.039+08:00 level=DEBUG source=sched.go:467 msg="context for request finished" time=2025-02-21T23:18:45.039+08:00 level=DEBUG source=sched.go:340 msg="runner with non-zero duration has gone idle, adding timer" modelPath=D:\ollama\models\blobs\sha256-553aa261cfb6856c595c9fefdb5453b98fdef331bf2ca918a5e0a23aa254d022 duration=5m0s time=2025-02-21T23:18:45.039+08:00 level=DEBUG source=sched.go:358 msg="after processing request finished event" modelPath=D:\ollama\models\blobs\sha256-553aa261cfb6856c595c9fefdb5453b98fdef331bf2ca918a5e0a23aa254d022 refCount=0 time=2025-02-21T23:18:54.108+08:00 level=DEBUG source=sched.go:576 msg="evaluating already loaded" model=D:\ollama\models\blobs\sha256-553aa261cfb6856c595c9fefdb5453b98fdef331bf2ca918a5e0a23aa254d022 time=2025-02-21T23:18:54.108+08:00 level=DEBUG source=routes.go:1462 msg="chat request" images=0 prompt=hi time=2025-02-21T23:18:54.108+08:00 level=DEBUG source=cache.go:104 msg="loading cache slot" id=0 cache=0 prompt=2 used=0 remaining=2
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#31784
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?