github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 16:11:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #11421] Error: model runner has unexpectedly stopped, this may be due to resource limitations or an internal error, check ollama server logs for details #54053

New Issue
Open
opened 2026-04-29 05:09:34 -05:00 by GiteaMirror · 10 comments
GiteaMirror commented 2026-04-29 05:09:34 -05:00
Owner
Copy Link

Originally created by @tts2 on GitHub (Jul 15, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/11421

What is the issue?

Windows 11 24H2
AMD Ryzen 7 PRO 8845HS w/ Radeon 780M Graphics 3.80 GHz
32.0 GB

Relevant log output

time=2025-07-15T09:31:26.800+08:00 level=INFO source=routes.go:1235 msg="server config" env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:4096 OLLAMA_DEBUG:INFO OLLAMA_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:C:\\Users\\tts2\\.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://* vscode-file://*] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES:]"
time=2025-07-15T09:31:26.810+08:00 level=INFO source=images.go:476 msg="total blobs: 8"
time=2025-07-15T09:31:26.811+08:00 level=INFO source=images.go:483 msg="total unused blobs removed: 0"
time=2025-07-15T09:31:26.811+08:00 level=INFO source=routes.go:1288 msg="Listening on [::]:11434 (version 0.9.6)"
time=2025-07-15T09:31:26.811+08:00 level=INFO source=gpu.go:217 msg="looking for compatible GPUs"
time=2025-07-15T09:31:26.811+08:00 level=INFO source=gpu_windows.go:167 msg=packages count=1
time=2025-07-15T09:31:26.811+08:00 level=INFO source=gpu_windows.go:214 msg="" package=0 cores=8 efficiency=0 threads=16
time=2025-07-15T09:31:26.878+08:00 level=INFO source=gpu.go:272 msg="error looking up nvidia GPU memory" error="cuda driver library failed to get device context 801"
time=2025-07-15T09:31:27.217+08:00 level=WARN source=amd_windows.go:138 msg="amdgpu is not supported (supported types:[gfx1030 gfx1100 gfx1101 gfx1102 gfx1151 gfx906])" gpu_type=gfx1103 gpu=0 library=C:\Users\tts2\AppData\Local\Programs\Ollama\lib\ollama\rocm
time=2025-07-15T09:31:27.217+08:00 level=INFO source=gpu.go:377 msg="no compatible GPUs were discovered"
time=2025-07-15T09:31:27.220+08:00 level=INFO source=types.go:130 msg="inference compute" id=0 library=cpu variant="" compute="" driver=0.0 name="" total="31.3 GiB" available="20.9 GiB"
time=2025-07-15T09:31:37.210+08:00 level=INFO source=server.go:135 msg="system memory" total="31.3 GiB" free="20.7 GiB" free_swap="17.4 GiB"
time=2025-07-15T09:31:37.211+08:00 level=INFO source=server.go:175 msg=offload library=cpu layers.requested=-1 layers.model=29 layers.offload=0 layers.split="" memory.available="[20.7 GiB]" memory.gpu_overhead="0 B" memory.required.full="2.3 GiB" memory.required.partial="0 B" memory.required.kv="896.0 MiB" memory.required.allocations="[2.3 GiB]" memory.weights.total="1.1 GiB" memory.weights.repeating="880.3 MiB" memory.weights.nonrepeating="243.4 MiB" memory.graph.full="298.7 MiB" memory.graph.partial="298.7 MiB"
llama_model_loader: loaded meta data with 27 key-value pairs and 311 tensors from C:\Users\tts2\.ollama\models\blobs\sha256-3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6 (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              = qwen3
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                               general.name str              = Qwen3 1.7B
llama_model_loader: - kv   3:                           general.basename str              = Qwen3
llama_model_loader: - kv   4:                         general.size_label str              = 1.7B
llama_model_loader: - kv   5:                          qwen3.block_count u32              = 28
llama_model_loader: - kv   6:                       qwen3.context_length u32              = 40960
llama_model_loader: - kv   7:                     qwen3.embedding_length u32              = 2048
llama_model_loader: - kv   8:                  qwen3.feed_forward_length u32              = 6144
llama_model_loader: - kv   9:                 qwen3.attention.head_count u32              = 16
llama_model_loader: - kv  10:              qwen3.attention.head_count_kv u32              = 8
llama_model_loader: - kv  11:                       qwen3.rope.freq_base f32              = 1000000.000000
llama_model_loader: - kv  12:     qwen3.attention.layer_norm_rms_epsilon f32              = 0.000001
llama_model_loader: - kv  13:                 qwen3.attention.key_length u32              = 128
llama_model_loader: - kv  14:               qwen3.attention.value_length u32              = 128
llama_model_loader: - kv  15:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  16:                         tokenizer.ggml.pre str              = qwen2
llama_model_loader: - kv  17:                      tokenizer.ggml.tokens arr[str,151936]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  18:                  tokenizer.ggml.token_type arr[i32,151936]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  19:                      tokenizer.ggml.merges arr[str,151387]  = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv  20:                tokenizer.ggml.eos_token_id u32              = 151645
llama_model_loader: - kv  21:            tokenizer.ggml.padding_token_id u32              = 151643
llama_model_loader: - kv  22:                tokenizer.ggml.bos_token_id u32              = 151643
llama_model_loader: - kv  23:               tokenizer.ggml.add_bos_token bool             = false
llama_model_loader: - kv  24:                    tokenizer.chat_template str              = {%- if tools %}\n    {{- '<|im_start|>...
llama_model_loader: - kv  25:               general.quantization_version u32              = 2
llama_model_loader: - kv  26:                          general.file_type u32              = 15
llama_model_loader: - type  f32:  113 tensors
llama_model_loader: - type  f16:   28 tensors
llama_model_loader: - type q4_K:  155 tensors
llama_model_loader: - type q6_K:   15 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q4_K - Medium
print_info: file size   = 1.26 GiB (5.33 BPW) 
load: special tokens cache size = 26
load: token to piece cache size = 0.9311 MB
print_info: arch             = qwen3
print_info: vocab_only       = 1
print_info: model type       = ?B
print_info: model params     = 2.03 B
print_info: general.name     = Qwen3 1.7B
print_info: vocab type       = BPE
print_info: n_vocab          = 151936
print_info: n_merges         = 151387
print_info: BOS token        = 151643 '<|endoftext|>'
print_info: EOS token        = 151645 '<|im_end|>'
print_info: EOT token        = 151645 '<|im_end|>'
print_info: PAD token        = 151643 '<|endoftext|>'
print_info: LF token         = 198 'Ċ'
print_info: FIM PRE token    = 151659 '<|fim_prefix|>'
print_info: FIM SUF token    = 151661 '<|fim_suffix|>'
print_info: FIM MID token    = 151660 '<|fim_middle|>'
print_info: FIM PAD token    = 151662 '<|fim_pad|>'
print_info: FIM REP token    = 151663 '<|repo_name|>'
print_info: FIM SEP token    = 151664 '<|file_sep|>'
print_info: EOG token        = 151643 '<|endoftext|>'
print_info: EOG token        = 151645 '<|im_end|>'
print_info: EOG token        = 151662 '<|fim_pad|>'
print_info: EOG token        = 151663 '<|repo_name|>'
print_info: EOG token        = 151664 '<|file_sep|>'
print_info: max token length = 256
llama_model_load: vocab only - skipping tensors
time=2025-07-15T09:31:37.430+08:00 level=INFO source=server.go:438 msg="starting llama server" cmd="C:\\Users\\tts2\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --model C:\\Users\\tts2\\.ollama\\models\\blobs\\sha256-3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6 --ctx-size 8192 --batch-size 512 --threads 8 --no-mmap --parallel 2 --port 59111"
time=2025-07-15T09:31:37.442+08:00 level=INFO source=sched.go:483 msg="loaded runners" count=1
time=2025-07-15T09:31:37.442+08:00 level=INFO source=server.go:598 msg="waiting for llama runner to start responding"
time=2025-07-15T09:31:37.442+08:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server error"
time=2025-07-15T09:31:37.476+08:00 level=INFO source=runner.go:815 msg="starting go runner"
ggml_cuda_init: GGML_CUDA_FORCE_MMQ:    no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 1 CUDA devices:
  Device 0: AMD Radeon 780M Graphics [ZLUDA], compute capability 8.8, VMM: no
load_backend: loaded CUDA backend from C:\Users\tts2\AppData\Local\Programs\Ollama\lib\ollama\ggml-cuda.dll
load_backend: loaded CPU backend from C:\Users\tts2\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-icelake.dll
time=2025-07-15T09:31:37.585+08:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.AVX512=1 CPU.0.AVX512_VBMI=1 CPU.0.AVX512_VNNI=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 CUDA.0.ARCHS=500,600,610,700,750,800,860,870,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(clang)
time=2025-07-15T09:31:37.586+08:00 level=INFO source=runner.go:874 msg="Server listening on 127.0.0.1:59111"
time=2025-07-15T09:31:37.693+08:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server loading model"
llama_model_load_from_file_impl: using device CUDA0 (AMD Radeon 780M Graphics [ZLUDA]) - 12391 MiB free
llama_model_loader: loaded meta data with 27 key-value pairs and 311 tensors from C:\Users\tts2\.ollama\models\blobs\sha256-3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6 (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              = qwen3
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                               general.name str              = Qwen3 1.7B
llama_model_loader: - kv   3:                           general.basename str              = Qwen3
llama_model_loader: - kv   4:                         general.size_label str              = 1.7B
llama_model_loader: - kv   5:                          qwen3.block_count u32              = 28
llama_model_loader: - kv   6:                       qwen3.context_length u32              = 40960
llama_model_loader: - kv   7:                     qwen3.embedding_length u32              = 2048
llama_model_loader: - kv   8:                  qwen3.feed_forward_length u32              = 6144
llama_model_loader: - kv   9:                 qwen3.attention.head_count u32              = 16
llama_model_loader: - kv  10:              qwen3.attention.head_count_kv u32              = 8
llama_model_loader: - kv  11:                       qwen3.rope.freq_base f32              = 1000000.000000
llama_model_loader: - kv  12:     qwen3.attention.layer_norm_rms_epsilon f32              = 0.000001
llama_model_loader: - kv  13:                 qwen3.attention.key_length u32              = 128
llama_model_loader: - kv  14:               qwen3.attention.value_length u32              = 128
llama_model_loader: - kv  15:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  16:                         tokenizer.ggml.pre str              = qwen2
llama_model_loader: - kv  17:                      tokenizer.ggml.tokens arr[str,151936]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  18:                  tokenizer.ggml.token_type arr[i32,151936]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  19:                      tokenizer.ggml.merges arr[str,151387]  = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv  20:                tokenizer.ggml.eos_token_id u32              = 151645
llama_model_loader: - kv  21:            tokenizer.ggml.padding_token_id u32              = 151643
llama_model_loader: - kv  22:                tokenizer.ggml.bos_token_id u32              = 151643
llama_model_loader: - kv  23:               tokenizer.ggml.add_bos_token bool             = false
llama_model_loader: - kv  24:                    tokenizer.chat_template str              = {%- if tools %}\n    {{- '<|im_start|>...
llama_model_loader: - kv  25:               general.quantization_version u32              = 2
llama_model_loader: - kv  26:                          general.file_type u32              = 15
llama_model_loader: - type  f32:  113 tensors
llama_model_loader: - type  f16:   28 tensors
llama_model_loader: - type q4_K:  155 tensors
llama_model_loader: - type q6_K:   15 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q4_K - Medium
print_info: file size   = 1.26 GiB (5.33 BPW) 
load: special tokens cache size = 26
load: token to piece cache size = 0.9311 MB
print_info: arch             = qwen3
print_info: vocab_only       = 0
print_info: n_ctx_train      = 40960
print_info: n_embd           = 2048
print_info: n_layer          = 28
print_info: n_head           = 16
print_info: n_head_kv        = 8
print_info: n_rot            = 128
print_info: n_swa            = 0
print_info: n_swa_pattern    = 1
print_info: n_embd_head_k    = 128
print_info: n_embd_head_v    = 128
print_info: n_gqa            = 2
print_info: n_embd_k_gqa     = 1024
print_info: n_embd_v_gqa     = 1024
print_info: f_norm_eps       = 0.0e+00
print_info: f_norm_rms_eps   = 1.0e-06
print_info: f_clamp_kqv      = 0.0e+00
print_info: f_max_alibi_bias = 0.0e+00
print_info: f_logit_scale    = 0.0e+00
print_info: f_attn_scale     = 0.0e+00
print_info: n_ff             = 6144
print_info: n_expert         = 0
print_info: n_expert_used    = 0
print_info: causal attn      = 1
print_info: pooling type     = 0
print_info: rope type        = 2
print_info: rope scaling     = linear
print_info: freq_base_train  = 1000000.0
print_info: freq_scale_train = 1
print_info: n_ctx_orig_yarn  = 40960
print_info: rope_finetuned   = unknown
print_info: ssm_d_conv       = 0
print_info: ssm_d_inner      = 0
print_info: ssm_d_state      = 0
print_info: ssm_dt_rank      = 0
print_info: ssm_dt_b_c_rms   = 0
print_info: model type       = 1.7B
print_info: model params     = 2.03 B
print_info: general.name     = Qwen3 1.7B
print_info: vocab type       = BPE
print_info: n_vocab          = 151936
print_info: n_merges         = 151387
print_info: BOS token        = 151643 '<|endoftext|>'
print_info: EOS token        = 151645 '<|im_end|>'
print_info: EOT token        = 151645 '<|im_end|>'
print_info: PAD token        = 151643 '<|endoftext|>'
print_info: LF token         = 198 'Ċ'
print_info: FIM PRE token    = 151659 '<|fim_prefix|>'
print_info: FIM SUF token    = 151661 '<|fim_suffix|>'
print_info: FIM MID token    = 151660 '<|fim_middle|>'
print_info: FIM PAD token    = 151662 '<|fim_pad|>'
print_info: FIM REP token    = 151663 '<|repo_name|>'
print_info: FIM SEP token    = 151664 '<|file_sep|>'
print_info: EOG token        = 151643 '<|endoftext|>'
print_info: EOG token        = 151645 '<|im_end|>'
print_info: EOG token        = 151662 '<|fim_pad|>'
print_info: EOG token        = 151663 '<|repo_name|>'
print_info: EOG token        = 151664 '<|file_sep|>'
print_info: max token length = 256
load_tensors: loading model tensors, this can take a while... (mmap = false)
load_tensors: offloading 0 repeating layers to GPU
load_tensors: offloaded 0/29 layers to GPU
load_tensors:          CPU model buffer size =   166.92 MiB
load_tensors:    CUDA_Host model buffer size =  1123.71 MiB
llama_context: constructing llama_context
llama_context: n_seq_max     = 2
llama_context: n_ctx         = 8192
llama_context: n_ctx_per_seq = 4096
llama_context: n_batch       = 1024
llama_context: n_ubatch      = 512
llama_context: causal_attn   = 1
llama_context: flash_attn    = 0
llama_context: freq_base     = 1000000.0
llama_context: freq_scale    = 1
llama_context: n_ctx_per_seq (4096) < n_ctx_train (40960) -- the full capacity of the model will not be utilized
llama_context:        CPU  output buffer size =     1.17 MiB
llama_kv_cache_unified: kv_size = 8192, type_k = 'f16', type_v = 'f16', n_layer = 28, can_shift = 1, padding = 32
llama_kv_cache_unified:        CPU KV buffer size =   896.00 MiB
llama_kv_cache_unified: KV self size  =  896.00 MiB, K (f16):  448.00 MiB, V (f16):  448.00 MiB
llama_context:      CUDA0 compute buffer size =   570.93 MiB
llama_context:  CUDA_Host compute buffer size =    20.01 MiB
llama_context: graph nodes  = 1070
llama_context: graph splits = 368 (with bs=512), 1 (with bs=1)
time=2025-07-15T09:31:38.695+08:00 level=INFO source=server.go:637 msg="llama runner started in 1.25 seconds"
C:\a\ollama\ollama\ml\backend\ggml\ggml\src\ggml-cuda\ggml-cuda.cu:2641: GGML_ASSERT(stat == cudaSuccess) failed
time=2025-07-15T09:31:38.954+08:00 level=ERROR source=server.go:807 msg="post predict" error="Post \"http://127.0.0.1:59111/completion\": read tcp 127.0.0.1:59113->127.0.0.1:59111: wsarecv: An existing connection was forcibly closed by the remote host."
[GIN] 2025/07/15 - 09:31:38 | 500 |    1.8302656s | 192.168.149.120 | POST     "/v1/chat/completions"
time=2025-07-15T09:31:39.066+08:00 level=ERROR source=server.go:464 msg="llama runner terminated" error="exit status 0xc0000409"
time=2025-07-15T09:31:39.503+08:00 level=INFO source=server.go:135 msg="system memory" total="31.3 GiB" free="20.6 GiB" free_swap="17.3 GiB"
time=2025-07-15T09:31:39.503+08:00 level=INFO source=server.go:175 msg=offload library=cpu layers.requested=-1 layers.model=29 layers.offload=0 layers.split="" memory.available="[20.6 GiB]" memory.gpu_overhead="0 B" memory.required.full="2.3 GiB" memory.required.partial="0 B" memory.required.kv="896.0 MiB" memory.required.allocations="[2.3 GiB]" memory.weights.total="1.1 GiB" memory.weights.repeating="880.3 MiB" memory.weights.nonrepeating="243.4 MiB" memory.graph.full="298.7 MiB" memory.graph.partial="298.7 MiB"
llama_model_loader: loaded meta data with 27 key-value pairs and 311 tensors from C:\Users\tts2\.ollama\models\blobs\sha256-3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6 (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              = qwen3
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                               general.name str              = Qwen3 1.7B
llama_model_loader: - kv   3:                           general.basename str              = Qwen3
llama_model_loader: - kv   4:                         general.size_label str              = 1.7B
llama_model_loader: - kv   5:                          qwen3.block_count u32              = 28
llama_model_loader: - kv   6:                       qwen3.context_length u32              = 40960
llama_model_loader: - kv   7:                     qwen3.embedding_length u32              = 2048
llama_model_loader: - kv   8:                  qwen3.feed_forward_length u32              = 6144
llama_model_loader: - kv   9:                 qwen3.attention.head_count u32              = 16
llama_model_loader: - kv  10:              qwen3.attention.head_count_kv u32              = 8
llama_model_loader: - kv  11:                       qwen3.rope.freq_base f32              = 1000000.000000
llama_model_loader: - kv  12:     qwen3.attention.layer_norm_rms_epsilon f32              = 0.000001
llama_model_loader: - kv  13:                 qwen3.attention.key_length u32              = 128
llama_model_loader: - kv  14:               qwen3.attention.value_length u32              = 128
llama_model_loader: - kv  15:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  16:                         tokenizer.ggml.pre str              = qwen2
llama_model_loader: - kv  17:                      tokenizer.ggml.tokens arr[str,151936]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  18:                  tokenizer.ggml.token_type arr[i32,151936]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  19:                      tokenizer.ggml.merges arr[str,151387]  = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv  20:                tokenizer.ggml.eos_token_id u32              = 151645
llama_model_loader: - kv  21:            tokenizer.ggml.padding_token_id u32              = 151643
llama_model_loader: - kv  22:                tokenizer.ggml.bos_token_id u32              = 151643
llama_model_loader: - kv  23:               tokenizer.ggml.add_bos_token bool             = false
llama_model_loader: - kv  24:                    tokenizer.chat_template str              = {%- if tools %}\n    {{- '<|im_start|>...
llama_model_loader: - kv  25:               general.quantization_version u32              = 2
llama_model_loader: - kv  26:                          general.file_type u32              = 15
llama_model_loader: - type  f32:  113 tensors
llama_model_loader: - type  f16:   28 tensors
llama_model_loader: - type q4_K:  155 tensors
llama_model_loader: - type q6_K:   15 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q4_K - Medium
print_info: file size   = 1.26 GiB (5.33 BPW) 
load: special tokens cache size = 26
load: token to piece cache size = 0.9311 MB
print_info: arch             = qwen3
print_info: vocab_only       = 1
print_info: model type       = ?B
print_info: model params     = 2.03 B
print_info: general.name     = Qwen3 1.7B
print_info: vocab type       = BPE
print_info: n_vocab          = 151936
print_info: n_merges         = 151387
print_info: BOS token        = 151643 '<|endoftext|>'
print_info: EOS token        = 151645 '<|im_end|>'
print_info: EOT token        = 151645 '<|im_end|>'
print_info: PAD token        = 151643 '<|endoftext|>'
print_info: LF token         = 198 'Ċ'
print_info: FIM PRE token    = 151659 '<|fim_prefix|>'
print_info: FIM SUF token    = 151661 '<|fim_suffix|>'
print_info: FIM MID token    = 151660 '<|fim_middle|>'
print_info: FIM PAD token    = 151662 '<|fim_pad|>'
print_info: FIM REP token    = 151663 '<|repo_name|>'
print_info: FIM SEP token    = 151664 '<|file_sep|>'
print_info: EOG token        = 151643 '<|endoftext|>'
print_info: EOG token        = 151645 '<|im_end|>'
print_info: EOG token        = 151662 '<|fim_pad|>'
print_info: EOG token        = 151663 '<|repo_name|>'
print_info: EOG token        = 151664 '<|file_sep|>'
print_info: max token length = 256
llama_model_load: vocab only - skipping tensors
time=2025-07-15T09:31:39.726+08:00 level=INFO source=server.go:438 msg="starting llama server" cmd="C:\\Users\\tts2\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --model C:\\Users\\tts2\\.ollama\\models\\blobs\\sha256-3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6 --ctx-size 8192 --batch-size 512 --threads 8 --no-mmap --parallel 2 --port 59114"
time=2025-07-15T09:31:39.738+08:00 level=INFO source=sched.go:483 msg="loaded runners" count=1
time=2025-07-15T09:31:39.738+08:00 level=INFO source=server.go:598 msg="waiting for llama runner to start responding"
time=2025-07-15T09:31:39.738+08:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server error"
time=2025-07-15T09:31:39.772+08:00 level=INFO source=runner.go:815 msg="starting go runner"
ggml_cuda_init: GGML_CUDA_FORCE_MMQ:    no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 1 CUDA devices:
  Device 0: AMD Radeon 780M Graphics [ZLUDA], compute capability 8.8, VMM: no
load_backend: loaded CUDA backend from C:\Users\tts2\AppData\Local\Programs\Ollama\lib\ollama\ggml-cuda.dll
load_backend: loaded CPU backend from C:\Users\tts2\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-icelake.dll
time=2025-07-15T09:31:39.882+08:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.AVX512=1 CPU.0.AVX512_VBMI=1 CPU.0.AVX512_VNNI=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 CUDA.0.ARCHS=500,600,610,700,750,800,860,870,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(clang)
time=2025-07-15T09:31:39.883+08:00 level=INFO source=runner.go:874 msg="Server listening on 127.0.0.1:59114"
time=2025-07-15T09:31:39.990+08:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server loading model"
llama_model_load_from_file_impl: using device CUDA0 (AMD Radeon 780M Graphics [ZLUDA]) - 12391 MiB free
llama_model_loader: loaded meta data with 27 key-value pairs and 311 tensors from C:\Users\tts2\.ollama\models\blobs\sha256-3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6 (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              = qwen3
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                               general.name str              = Qwen3 1.7B
llama_model_loader: - kv   3:                           general.basename str              = Qwen3
llama_model_loader: - kv   4:                         general.size_label str              = 1.7B
llama_model_loader: - kv   5:                          qwen3.block_count u32              = 28
llama_model_loader: - kv   6:                       qwen3.context_length u32              = 40960
llama_model_loader: - kv   7:                     qwen3.embedding_length u32              = 2048
llama_model_loader: - kv   8:                  qwen3.feed_forward_length u32              = 6144
llama_model_loader: - kv   9:                 qwen3.attention.head_count u32              = 16
llama_model_loader: - kv  10:              qwen3.attention.head_count_kv u32              = 8
llama_model_loader: - kv  11:                       qwen3.rope.freq_base f32              = 1000000.000000
llama_model_loader: - kv  12:     qwen3.attention.layer_norm_rms_epsilon f32              = 0.000001
llama_model_loader: - kv  13:                 qwen3.attention.key_length u32              = 128
llama_model_loader: - kv  14:               qwen3.attention.value_length u32              = 128
llama_model_loader: - kv  15:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  16:                         tokenizer.ggml.pre str              = qwen2
llama_model_loader: - kv  17:                      tokenizer.ggml.tokens arr[str,151936]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  18:                  tokenizer.ggml.token_type arr[i32,151936]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  19:                      tokenizer.ggml.merges arr[str,151387]  = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv  20:                tokenizer.ggml.eos_token_id u32              = 151645
llama_model_loader: - kv  21:            tokenizer.ggml.padding_token_id u32              = 151643
llama_model_loader: - kv  22:                tokenizer.ggml.bos_token_id u32              = 151643
llama_model_loader: - kv  23:               tokenizer.ggml.add_bos_token bool             = false
llama_model_loader: - kv  24:                    tokenizer.chat_template str              = {%- if tools %}\n    {{- '<|im_start|>...
llama_model_loader: - kv  25:               general.quantization_version u32              = 2
llama_model_loader: - kv  26:                          general.file_type u32              = 15
llama_model_loader: - type  f32:  113 tensors
llama_model_loader: - type  f16:   28 tensors
llama_model_loader: - type q4_K:  155 tensors
llama_model_loader: - type q6_K:   15 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q4_K - Medium
print_info: file size   = 1.26 GiB (5.33 BPW) 
load: special tokens cache size = 26
load: token to piece cache size = 0.9311 MB
print_info: arch             = qwen3
print_info: vocab_only       = 0
print_info: n_ctx_train      = 40960
print_info: n_embd           = 2048
print_info: n_layer          = 28
print_info: n_head           = 16
print_info: n_head_kv        = 8
print_info: n_rot            = 128
print_info: n_swa            = 0
print_info: n_swa_pattern    = 1
print_info: n_embd_head_k    = 128
print_info: n_embd_head_v    = 128
print_info: n_gqa            = 2
print_info: n_embd_k_gqa     = 1024
print_info: n_embd_v_gqa     = 1024
print_info: f_norm_eps       = 0.0e+00
print_info: f_norm_rms_eps   = 1.0e-06
print_info: f_clamp_kqv      = 0.0e+00
print_info: f_max_alibi_bias = 0.0e+00
print_info: f_logit_scale    = 0.0e+00
print_info: f_attn_scale     = 0.0e+00
print_info: n_ff             = 6144
print_info: n_expert         = 0
print_info: n_expert_used    = 0
print_info: causal attn      = 1
print_info: pooling type     = 0
print_info: rope type        = 2
print_info: rope scaling     = linear
print_info: freq_base_train  = 1000000.0
print_info: freq_scale_train = 1
print_info: n_ctx_orig_yarn  = 40960
print_info: rope_finetuned   = unknown
print_info: ssm_d_conv       = 0
print_info: ssm_d_inner      = 0
print_info: ssm_d_state      = 0
print_info: ssm_dt_rank      = 0
print_info: ssm_dt_b_c_rms   = 0
print_info: model type       = 1.7B
print_info: model params     = 2.03 B
print_info: general.name     = Qwen3 1.7B
print_info: vocab type       = BPE
print_info: n_vocab          = 151936
print_info: n_merges         = 151387
print_info: BOS token        = 151643 '<|endoftext|>'
print_info: EOS token        = 151645 '<|im_end|>'
print_info: EOT token        = 151645 '<|im_end|>'
print_info: PAD token        = 151643 '<|endoftext|>'
print_info: LF token         = 198 'Ċ'
print_info: FIM PRE token    = 151659 '<|fim_prefix|>'
print_info: FIM SUF token    = 151661 '<|fim_suffix|>'
print_info: FIM MID token    = 151660 '<|fim_middle|>'
print_info: FIM PAD token    = 151662 '<|fim_pad|>'
print_info: FIM REP token    = 151663 '<|repo_name|>'
print_info: FIM SEP token    = 151664 '<|file_sep|>'
print_info: EOG token        = 151643 '<|endoftext|>'
print_info: EOG token        = 151645 '<|im_end|>'
print_info: EOG token        = 151662 '<|fim_pad|>'
print_info: EOG token        = 151663 '<|repo_name|>'
print_info: EOG token        = 151664 '<|file_sep|>'
print_info: max token length = 256
load_tensors: loading model tensors, this can take a while... (mmap = false)
load_tensors: offloading 0 repeating layers to GPU
load_tensors: offloaded 0/29 layers to GPU
load_tensors:          CPU model buffer size =   166.92 MiB
load_tensors:    CUDA_Host model buffer size =  1123.71 MiB
llama_context: constructing llama_context
llama_context: n_seq_max     = 2
llama_context: n_ctx         = 8192
llama_context: n_ctx_per_seq = 4096
llama_context: n_batch       = 1024
llama_context: n_ubatch      = 512
llama_context: causal_attn   = 1
llama_context: flash_attn    = 0
llama_context: freq_base     = 1000000.0
llama_context: freq_scale    = 1
llama_context: n_ctx_per_seq (4096) < n_ctx_train (40960) -- the full capacity of the model will not be utilized
llama_context:        CPU  output buffer size =     1.17 MiB
llama_kv_cache_unified: kv_size = 8192, type_k = 'f16', type_v = 'f16', n_layer = 28, can_shift = 1, padding = 32
llama_kv_cache_unified:        CPU KV buffer size =   896.00 MiB
llama_kv_cache_unified: KV self size  =  896.00 MiB, K (f16):  448.00 MiB, V (f16):  448.00 MiB
llama_context:      CUDA0 compute buffer size =   570.93 MiB
llama_context:  CUDA_Host compute buffer size =    20.01 MiB
llama_context: graph nodes  = 1070
llama_context: graph splits = 368 (with bs=512), 1 (with bs=1)
time=2025-07-15T09:31:40.991+08:00 level=INFO source=server.go:637 msg="llama runner started in 1.25 seconds"
C:\a\ollama\ollama\ml\backend\ggml\ggml\src\ggml-cuda\ggml-cuda.cu:2641: GGML_ASSERT(stat == cudaSuccess) failed
time=2025-07-15T09:31:41.197+08:00 level=ERROR source=server.go:807 msg="post predict" error="Post \"http://127.0.0.1:59114/completion\": read tcp 127.0.0.1:59116->127.0.0.1:59114: wsarecv: An existing connection was forcibly closed by the remote host."
[GIN] 2025/07/15 - 09:31:41 | 500 |    1.8284561s | 192.168.149.120 | POST     "/v1/chat/completions"
time=2025-07-15T09:31:41.304+08:00 level=ERROR source=server.go:464 msg="llama runner terminated" error="exit status 0xc0000409"
time=2025-07-15T09:31:42.173+08:00 level=INFO source=server.go:135 msg="system memory" total="31.3 GiB" free="20.6 GiB" free_swap="17.3 GiB"
time=2025-07-15T09:31:42.173+08:00 level=INFO source=server.go:175 msg=offload library=cpu layers.requested=-1 layers.model=29 layers.offload=0 layers.split="" memory.available="[20.6 GiB]" memory.gpu_overhead="0 B" memory.required.full="2.3 GiB" memory.required.partial="0 B" memory.required.kv="896.0 MiB" memory.required.allocations="[2.3 GiB]" memory.weights.total="1.1 GiB" memory.weights.repeating="880.3 MiB" memory.weights.nonrepeating="243.4 MiB" memory.graph.full="298.7 MiB" memory.graph.partial="298.7 MiB"
llama_model_loader: loaded meta data with 27 key-value pairs and 311 tensors from C:\Users\tts2\.ollama\models\blobs\sha256-3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6 (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              = qwen3
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                               general.name str              = Qwen3 1.7B
llama_model_loader: - kv   3:                           general.basename str              = Qwen3
llama_model_loader: - kv   4:                         general.size_label str              = 1.7B
llama_model_loader: - kv   5:                          qwen3.block_count u32              = 28
llama_model_loader: - kv   6:                       qwen3.context_length u32              = 40960
llama_model_loader: - kv   7:                     qwen3.embedding_length u32              = 2048
llama_model_loader: - kv   8:                  qwen3.feed_forward_length u32              = 6144
llama_model_loader: - kv   9:                 qwen3.attention.head_count u32              = 16
llama_model_loader: - kv  10:              qwen3.attention.head_count_kv u32              = 8
llama_model_loader: - kv  11:                       qwen3.rope.freq_base f32              = 1000000.000000
llama_model_loader: - kv  12:     qwen3.attention.layer_norm_rms_epsilon f32              = 0.000001
llama_model_loader: - kv  13:                 qwen3.attention.key_length u32              = 128
llama_model_loader: - kv  14:               qwen3.attention.value_length u32              = 128
llama_model_loader: - kv  15:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  16:                         tokenizer.ggml.pre str              = qwen2
llama_model_loader: - kv  17:                      tokenizer.ggml.tokens arr[str,151936]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  18:                  tokenizer.ggml.token_type arr[i32,151936]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  19:                      tokenizer.ggml.merges arr[str,151387]  = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv  20:                tokenizer.ggml.eos_token_id u32              = 151645
llama_model_loader: - kv  21:            tokenizer.ggml.padding_token_id u32              = 151643
llama_model_loader: - kv  22:                tokenizer.ggml.bos_token_id u32              = 151643
llama_model_loader: - kv  23:               tokenizer.ggml.add_bos_token bool             = false
llama_model_loader: - kv  24:                    tokenizer.chat_template str              = {%- if tools %}\n    {{- '<|im_start|>...
llama_model_loader: - kv  25:               general.quantization_version u32              = 2
llama_model_loader: - kv  26:                          general.file_type u32              = 15
llama_model_loader: - type  f32:  113 tensors
llama_model_loader: - type  f16:   28 tensors
llama_model_loader: - type q4_K:  155 tensors
llama_model_loader: - type q6_K:   15 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q4_K - Medium
print_info: file size   = 1.26 GiB (5.33 BPW) 
load: special tokens cache size = 26
load: token to piece cache size = 0.9311 MB
print_info: arch             = qwen3
print_info: vocab_only       = 1
print_info: model type       = ?B
print_info: model params     = 2.03 B
print_info: general.name     = Qwen3 1.7B
print_info: vocab type       = BPE
print_info: n_vocab          = 151936
print_info: n_merges         = 151387
print_info: BOS token        = 151643 '<|endoftext|>'
print_info: EOS token        = 151645 '<|im_end|>'
print_info: EOT token        = 151645 '<|im_end|>'
print_info: PAD token        = 151643 '<|endoftext|>'
print_info: LF token         = 198 'Ċ'
print_info: FIM PRE token    = 151659 '<|fim_prefix|>'
print_info: FIM SUF token    = 151661 '<|fim_suffix|>'
print_info: FIM MID token    = 151660 '<|fim_middle|>'
print_info: FIM PAD token    = 151662 '<|fim_pad|>'
print_info: FIM REP token    = 151663 '<|repo_name|>'
print_info: FIM SEP token    = 151664 '<|file_sep|>'
print_info: EOG token        = 151643 '<|endoftext|>'
print_info: EOG token        = 151645 '<|im_end|>'
print_info: EOG token        = 151662 '<|fim_pad|>'
print_info: EOG token        = 151663 '<|repo_name|>'
print_info: EOG token        = 151664 '<|file_sep|>'
print_info: max token length = 256
llama_model_load: vocab only - skipping tensors
time=2025-07-15T09:31:42.374+08:00 level=INFO source=server.go:438 msg="starting llama server" cmd="C:\\Users\\tts2\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --model C:\\Users\\tts2\\.ollama\\models\\blobs\\sha256-3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6 --ctx-size 8192 --batch-size 512 --threads 8 --no-mmap --parallel 2 --port 59117"
time=2025-07-15T09:31:42.386+08:00 level=INFO source=sched.go:483 msg="loaded runners" count=1
time=2025-07-15T09:31:42.386+08:00 level=INFO source=server.go:598 msg="waiting for llama runner to start responding"
time=2025-07-15T09:31:42.386+08:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server error"
time=2025-07-15T09:31:42.419+08:00 level=INFO source=runner.go:815 msg="starting go runner"
ggml_cuda_init: GGML_CUDA_FORCE_MMQ:    no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 1 CUDA devices:
  Device 0: AMD Radeon 780M Graphics [ZLUDA], compute capability 8.8, VMM: no
load_backend: loaded CUDA backend from C:\Users\tts2\AppData\Local\Programs\Ollama\lib\ollama\ggml-cuda.dll
load_backend: loaded CPU backend from C:\Users\tts2\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-icelake.dll
time=2025-07-15T09:31:42.530+08:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.AVX512=1 CPU.0.AVX512_VBMI=1 CPU.0.AVX512_VNNI=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 CUDA.0.ARCHS=500,600,610,700,750,800,860,870,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(clang)
time=2025-07-15T09:31:42.530+08:00 level=INFO source=runner.go:874 msg="Server listening on 127.0.0.1:59117"
time=2025-07-15T09:31:42.637+08:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server loading model"
llama_model_load_from_file_impl: using device CUDA0 (AMD Radeon 780M Graphics [ZLUDA]) - 12391 MiB free
llama_model_loader: loaded meta data with 27 key-value pairs and 311 tensors from C:\Users\tts2\.ollama\models\blobs\sha256-3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6 (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              = qwen3
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                               general.name str              = Qwen3 1.7B
llama_model_loader: - kv   3:                           general.basename str              = Qwen3
llama_model_loader: - kv   4:                         general.size_label str              = 1.7B
llama_model_loader: - kv   5:                          qwen3.block_count u32              = 28
llama_model_loader: - kv   6:                       qwen3.context_length u32              = 40960
llama_model_loader: - kv   7:                     qwen3.embedding_length u32              = 2048
llama_model_loader: - kv   8:                  qwen3.feed_forward_length u32              = 6144
llama_model_loader: - kv   9:                 qwen3.attention.head_count u32              = 16
llama_model_loader: - kv  10:              qwen3.attention.head_count_kv u32              = 8
llama_model_loader: - kv  11:                       qwen3.rope.freq_base f32              = 1000000.000000
llama_model_loader: - kv  12:     qwen3.attention.layer_norm_rms_epsilon f32              = 0.000001
llama_model_loader: - kv  13:                 qwen3.attention.key_length u32              = 128
llama_model_loader: - kv  14:               qwen3.attention.value_length u32              = 128
llama_model_loader: - kv  15:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  16:                         tokenizer.ggml.pre str              = qwen2
llama_model_loader: - kv  17:                      tokenizer.ggml.tokens arr[str,151936]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  18:                  tokenizer.ggml.token_type arr[i32,151936]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  19:                      tokenizer.ggml.merges arr[str,151387]  = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv  20:                tokenizer.ggml.eos_token_id u32              = 151645
llama_model_loader: - kv  21:            tokenizer.ggml.padding_token_id u32              = 151643
llama_model_loader: - kv  22:                tokenizer.ggml.bos_token_id u32              = 151643
llama_model_loader: - kv  23:               tokenizer.ggml.add_bos_token bool             = false
llama_model_loader: - kv  24:                    tokenizer.chat_template str              = {%- if tools %}\n    {{- '<|im_start|>...
llama_model_loader: - kv  25:               general.quantization_version u32              = 2
llama_model_loader: - kv  26:                          general.file_type u32              = 15
llama_model_loader: - type  f32:  113 tensors
llama_model_loader: - type  f16:   28 tensors
llama_model_loader: - type q4_K:  155 tensors
llama_model_loader: - type q6_K:   15 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q4_K - Medium
print_info: file size   = 1.26 GiB (5.33 BPW) 
load: special tokens cache size = 26
load: token to piece cache size = 0.9311 MB
print_info: arch             = qwen3
print_info: vocab_only       = 0
print_info: n_ctx_train      = 40960
print_info: n_embd           = 2048
print_info: n_layer          = 28
print_info: n_head           = 16
print_info: n_head_kv        = 8
print_info: n_rot            = 128
print_info: n_swa            = 0
print_info: n_swa_pattern    = 1
print_info: n_embd_head_k    = 128
print_info: n_embd_head_v    = 128
print_info: n_gqa            = 2
print_info: n_embd_k_gqa     = 1024
print_info: n_embd_v_gqa     = 1024
print_info: f_norm_eps       = 0.0e+00
print_info: f_norm_rms_eps   = 1.0e-06
print_info: f_clamp_kqv      = 0.0e+00
print_info: f_max_alibi_bias = 0.0e+00
print_info: f_logit_scale    = 0.0e+00
print_info: f_attn_scale     = 0.0e+00
print_info: n_ff             = 6144
print_info: n_expert         = 0
print_info: n_expert_used    = 0
print_info: causal attn      = 1
print_info: pooling type     = 0
print_info: rope type        = 2
print_info: rope scaling     = linear
print_info: freq_base_train  = 1000000.0
print_info: freq_scale_train = 1
print_info: n_ctx_orig_yarn  = 40960
print_info: rope_finetuned   = unknown
print_info: ssm_d_conv       = 0
print_info: ssm_d_inner      = 0
print_info: ssm_d_state      = 0
print_info: ssm_dt_rank      = 0
print_info: ssm_dt_b_c_rms   = 0
print_info: model type       = 1.7B
print_info: model params     = 2.03 B
print_info: general.name     = Qwen3 1.7B
print_info: vocab type       = BPE
print_info: n_vocab          = 151936
print_info: n_merges         = 151387
print_info: BOS token        = 151643 '<|endoftext|>'
print_info: EOS token        = 151645 '<|im_end|>'
print_info: EOT token        = 151645 '<|im_end|>'
print_info: PAD token        = 151643 '<|endoftext|>'
print_info: LF token         = 198 'Ċ'
print_info: FIM PRE token    = 151659 '<|fim_prefix|>'
print_info: FIM SUF token    = 151661 '<|fim_suffix|>'
print_info: FIM MID token    = 151660 '<|fim_middle|>'
print_info: FIM PAD token    = 151662 '<|fim_pad|>'
print_info: FIM REP token    = 151663 '<|repo_name|>'
print_info: FIM SEP token    = 151664 '<|file_sep|>'
print_info: EOG token        = 151643 '<|endoftext|>'
print_info: EOG token        = 151645 '<|im_end|>'
print_info: EOG token        = 151662 '<|fim_pad|>'
print_info: EOG token        = 151663 '<|repo_name|>'
print_info: EOG token        = 151664 '<|file_sep|>'
print_info: max token length = 256
load_tensors: loading model tensors, this can take a while... (mmap = false)
load_tensors: offloading 0 repeating layers to GPU
load_tensors: offloaded 0/29 layers to GPU
load_tensors:          CPU model buffer size =   166.92 MiB
load_tensors:    CUDA_Host model buffer size =  1123.71 MiB
llama_context: constructing llama_context
llama_context: n_seq_max     = 2
llama_context: n_ctx         = 8192
llama_context: n_ctx_per_seq = 4096
llama_context: n_batch       = 1024
llama_context: n_ubatch      = 512
llama_context: causal_attn   = 1
llama_context: flash_attn    = 0
llama_context: freq_base     = 1000000.0
llama_context: freq_scale    = 1
llama_context: n_ctx_per_seq (4096) < n_ctx_train (40960) -- the full capacity of the model will not be utilized
llama_context:        CPU  output buffer size =     1.17 MiB
llama_kv_cache_unified: kv_size = 8192, type_k = 'f16', type_v = 'f16', n_layer = 28, can_shift = 1, padding = 32
llama_kv_cache_unified:        CPU KV buffer size =   896.00 MiB
llama_kv_cache_unified: KV self size  =  896.00 MiB, K (f16):  448.00 MiB, V (f16):  448.00 MiB
llama_context:      CUDA0 compute buffer size =   570.93 MiB
llama_context:  CUDA_Host compute buffer size =    20.01 MiB
llama_context: graph nodes  = 1070
llama_context: graph splits = 368 (with bs=512), 1 (with bs=1)
time=2025-07-15T09:31:43.639+08:00 level=INFO source=server.go:637 msg="llama runner started in 1.25 seconds"
C:\a\ollama\ollama\ml\backend\ggml\ggml\src\ggml-cuda\ggml-cuda.cu:2641: GGML_ASSERT(stat == cudaSuccess) failed
time=2025-07-15T09:31:43.838+08:00 level=ERROR source=server.go:807 msg="post predict" error="Post \"http://127.0.0.1:59117/completion\": read tcp 127.0.0.1:59119->127.0.0.1:59117: wsarecv: An existing connection was forcibly closed by the remote host."
[GIN] 2025/07/15 - 09:31:43 | 500 |    1.7839638s | 192.168.149.120 | POST     "/v1/chat/completions"
time=2025-07-15T09:31:43.938+08:00 level=ERROR source=server.go:464 msg="llama runner terminated" error="exit status 0xc0000409"
[GIN] 2025/07/15 - 09:46:10 | 200 |       578.8µs |       127.0.0.1 | HEAD     "/"
[GIN] 2025/07/15 - 09:46:10 | 200 |      55.609ms |       127.0.0.1 | POST     "/api/show"
time=2025-07-15T09:46:11.074+08:00 level=INFO source=server.go:135 msg="system memory" total="31.3 GiB" free="20.5 GiB" free_swap="17.4 GiB"
time=2025-07-15T09:46:11.074+08:00 level=INFO source=server.go:175 msg=offload library=cpu layers.requested=-1 layers.model=29 layers.offload=0 layers.split="" memory.available="[20.5 GiB]" memory.gpu_overhead="0 B" memory.required.full="2.3 GiB" memory.required.partial="0 B" memory.required.kv="896.0 MiB" memory.required.allocations="[2.3 GiB]" memory.weights.total="1.1 GiB" memory.weights.repeating="880.3 MiB" memory.weights.nonrepeating="243.4 MiB" memory.graph.full="298.7 MiB" memory.graph.partial="298.7 MiB"
llama_model_loader: loaded meta data with 27 key-value pairs and 311 tensors from C:\Users\tts2\.ollama\models\blobs\sha256-3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6 (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              = qwen3
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                               general.name str              = Qwen3 1.7B
llama_model_loader: - kv   3:                           general.basename str              = Qwen3
llama_model_loader: - kv   4:                         general.size_label str              = 1.7B
llama_model_loader: - kv   5:                          qwen3.block_count u32              = 28
llama_model_loader: - kv   6:                       qwen3.context_length u32              = 40960
llama_model_loader: - kv   7:                     qwen3.embedding_length u32              = 2048
llama_model_loader: - kv   8:                  qwen3.feed_forward_length u32              = 6144
llama_model_loader: - kv   9:                 qwen3.attention.head_count u32              = 16
llama_model_loader: - kv  10:              qwen3.attention.head_count_kv u32              = 8
llama_model_loader: - kv  11:                       qwen3.rope.freq_base f32              = 1000000.000000
llama_model_loader: - kv  12:     qwen3.attention.layer_norm_rms_epsilon f32              = 0.000001
llama_model_loader: - kv  13:                 qwen3.attention.key_length u32              = 128
llama_model_loader: - kv  14:               qwen3.attention.value_length u32              = 128
llama_model_loader: - kv  15:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  16:                         tokenizer.ggml.pre str              = qwen2
llama_model_loader: - kv  17:                      tokenizer.ggml.tokens arr[str,151936]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  18:                  tokenizer.ggml.token_type arr[i32,151936]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  19:                      tokenizer.ggml.merges arr[str,151387]  = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv  20:                tokenizer.ggml.eos_token_id u32              = 151645
llama_model_loader: - kv  21:            tokenizer.ggml.padding_token_id u32              = 151643
llama_model_loader: - kv  22:                tokenizer.ggml.bos_token_id u32              = 151643
llama_model_loader: - kv  23:               tokenizer.ggml.add_bos_token bool             = false
llama_model_loader: - kv  24:                    tokenizer.chat_template str              = {%- if tools %}\n    {{- '<|im_start|>...
llama_model_loader: - kv  25:               general.quantization_version u32              = 2
llama_model_loader: - kv  26:                          general.file_type u32              = 15
llama_model_loader: - type  f32:  113 tensors
llama_model_loader: - type  f16:   28 tensors
llama_model_loader: - type q4_K:  155 tensors
llama_model_loader: - type q6_K:   15 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q4_K - Medium
print_info: file size   = 1.26 GiB (5.33 BPW) 
load: special tokens cache size = 26
load: token to piece cache size = 0.9311 MB
print_info: arch             = qwen3
print_info: vocab_only       = 1
print_info: model type       = ?B
print_info: model params     = 2.03 B
print_info: general.name     = Qwen3 1.7B
print_info: vocab type       = BPE
print_info: n_vocab          = 151936
print_info: n_merges         = 151387
print_info: BOS token        = 151643 '<|endoftext|>'
print_info: EOS token        = 151645 '<|im_end|>'
print_info: EOT token        = 151645 '<|im_end|>'
print_info: PAD token        = 151643 '<|endoftext|>'
print_info: LF token         = 198 'Ċ'
print_info: FIM PRE token    = 151659 '<|fim_prefix|>'
print_info: FIM SUF token    = 151661 '<|fim_suffix|>'
print_info: FIM MID token    = 151660 '<|fim_middle|>'
print_info: FIM PAD token    = 151662 '<|fim_pad|>'
print_info: FIM REP token    = 151663 '<|repo_name|>'
print_info: FIM SEP token    = 151664 '<|file_sep|>'
print_info: EOG token        = 151643 '<|endoftext|>'
print_info: EOG token        = 151645 '<|im_end|>'
print_info: EOG token        = 151662 '<|fim_pad|>'
print_info: EOG token        = 151663 '<|repo_name|>'
print_info: EOG token        = 151664 '<|file_sep|>'
print_info: max token length = 256
llama_model_load: vocab only - skipping tensors
time=2025-07-15T09:46:11.303+08:00 level=INFO source=server.go:438 msg="starting llama server" cmd="C:\\Users\\tts2\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --model C:\\Users\\tts2\\.ollama\\models\\blobs\\sha256-3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6 --ctx-size 8192 --batch-size 512 --threads 8 --no-mmap --parallel 2 --port 59254"
time=2025-07-15T09:46:11.317+08:00 level=INFO source=sched.go:483 msg="loaded runners" count=1
time=2025-07-15T09:46:11.317+08:00 level=INFO source=server.go:598 msg="waiting for llama runner to start responding"
time=2025-07-15T09:46:11.318+08:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server error"
time=2025-07-15T09:46:11.357+08:00 level=INFO source=runner.go:815 msg="starting go runner"
ggml_cuda_init: GGML_CUDA_FORCE_MMQ:    no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 1 CUDA devices:
  Device 0: AMD Radeon 780M Graphics [ZLUDA], compute capability 8.8, VMM: no
load_backend: loaded CUDA backend from C:\Users\tts2\AppData\Local\Programs\Ollama\lib\ollama\ggml-cuda.dll
load_backend: loaded CPU backend from C:\Users\tts2\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-icelake.dll
time=2025-07-15T09:46:11.483+08:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.AVX512=1 CPU.0.AVX512_VBMI=1 CPU.0.AVX512_VNNI=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 CUDA.0.ARCHS=500,600,610,700,750,800,860,870,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(clang)
time=2025-07-15T09:46:11.484+08:00 level=INFO source=runner.go:874 msg="Server listening on 127.0.0.1:59254"
time=2025-07-15T09:46:11.569+08:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server loading model"
llama_model_load_from_file_impl: using device CUDA0 (AMD Radeon 780M Graphics [ZLUDA]) - 12391 MiB free
llama_model_loader: loaded meta data with 27 key-value pairs and 311 tensors from C:\Users\tts2\.ollama\models\blobs\sha256-3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6 (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              = qwen3
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                               general.name str              = Qwen3 1.7B
llama_model_loader: - kv   3:                           general.basename str              = Qwen3
llama_model_loader: - kv   4:                         general.size_label str              = 1.7B
llama_model_loader: - kv   5:                          qwen3.block_count u32              = 28
llama_model_loader: - kv   6:                       qwen3.context_length u32              = 40960
llama_model_loader: - kv   7:                     qwen3.embedding_length u32              = 2048
llama_model_loader: - kv   8:                  qwen3.feed_forward_length u32              = 6144
llama_model_loader: - kv   9:                 qwen3.attention.head_count u32              = 16
llama_model_loader: - kv  10:              qwen3.attention.head_count_kv u32              = 8
llama_model_loader: - kv  11:                       qwen3.rope.freq_base f32              = 1000000.000000
llama_model_loader: - kv  12:     qwen3.attention.layer_norm_rms_epsilon f32              = 0.000001
llama_model_loader: - kv  13:                 qwen3.attention.key_length u32              = 128
llama_model_loader: - kv  14:               qwen3.attention.value_length u32              = 128
llama_model_loader: - kv  15:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  16:                         tokenizer.ggml.pre str              = qwen2
llama_model_loader: - kv  17:                      tokenizer.ggml.tokens arr[str,151936]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  18:                  tokenizer.ggml.token_type arr[i32,151936]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  19:                      tokenizer.ggml.merges arr[str,151387]  = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv  20:                tokenizer.ggml.eos_token_id u32              = 151645
llama_model_loader: - kv  21:            tokenizer.ggml.padding_token_id u32              = 151643
llama_model_loader: - kv  22:                tokenizer.ggml.bos_token_id u32              = 151643
llama_model_loader: - kv  23:               tokenizer.ggml.add_bos_token bool             = false
llama_model_loader: - kv  24:                    tokenizer.chat_template str              = {%- if tools %}\n    {{- '<|im_start|>...
llama_model_loader: - kv  25:               general.quantization_version u32              = 2
llama_model_loader: - kv  26:                          general.file_type u32              = 15
llama_model_loader: - type  f32:  113 tensors
llama_model_loader: - type  f16:   28 tensors
llama_model_loader: - type q4_K:  155 tensors
llama_model_loader: - type q6_K:   15 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q4_K - Medium
print_info: file size   = 1.26 GiB (5.33 BPW) 
load: special tokens cache size = 26
load: token to piece cache size = 0.9311 MB
print_info: arch             = qwen3
print_info: vocab_only       = 0
print_info: n_ctx_train      = 40960
print_info: n_embd           = 2048
print_info: n_layer          = 28
print_info: n_head           = 16
print_info: n_head_kv        = 8
print_info: n_rot            = 128
print_info: n_swa            = 0
print_info: n_swa_pattern    = 1
print_info: n_embd_head_k    = 128
print_info: n_embd_head_v    = 128
print_info: n_gqa            = 2
print_info: n_embd_k_gqa     = 1024
print_info: n_embd_v_gqa     = 1024
print_info: f_norm_eps       = 0.0e+00
print_info: f_norm_rms_eps   = 1.0e-06
print_info: f_clamp_kqv      = 0.0e+00
print_info: f_max_alibi_bias = 0.0e+00
print_info: f_logit_scale    = 0.0e+00
print_info: f_attn_scale     = 0.0e+00
print_info: n_ff             = 6144
print_info: n_expert         = 0
print_info: n_expert_used    = 0
print_info: causal attn      = 1
print_info: pooling type     = 0
print_info: rope type        = 2
print_info: rope scaling     = linear
print_info: freq_base_train  = 1000000.0
print_info: freq_scale_train = 1
print_info: n_ctx_orig_yarn  = 40960
print_info: rope_finetuned   = unknown
print_info: ssm_d_conv       = 0
print_info: ssm_d_inner      = 0
print_info: ssm_d_state      = 0
print_info: ssm_dt_rank      = 0
print_info: ssm_dt_b_c_rms   = 0
print_info: model type       = 1.7B
print_info: model params     = 2.03 B
print_info: general.name     = Qwen3 1.7B
print_info: vocab type       = BPE
print_info: n_vocab          = 151936
print_info: n_merges         = 151387
print_info: BOS token        = 151643 '<|endoftext|>'
print_info: EOS token        = 151645 '<|im_end|>'
print_info: EOT token        = 151645 '<|im_end|>'
print_info: PAD token        = 151643 '<|endoftext|>'
print_info: LF token         = 198 'Ċ'
print_info: FIM PRE token    = 151659 '<|fim_prefix|>'
print_info: FIM SUF token    = 151661 '<|fim_suffix|>'
print_info: FIM MID token    = 151660 '<|fim_middle|>'
print_info: FIM PAD token    = 151662 '<|fim_pad|>'
print_info: FIM REP token    = 151663 '<|repo_name|>'
print_info: FIM SEP token    = 151664 '<|file_sep|>'
print_info: EOG token        = 151643 '<|endoftext|>'
print_info: EOG token        = 151645 '<|im_end|>'
print_info: EOG token        = 151662 '<|fim_pad|>'
print_info: EOG token        = 151663 '<|repo_name|>'
print_info: EOG token        = 151664 '<|file_sep|>'
print_info: max token length = 256
load_tensors: loading model tensors, this can take a while... (mmap = false)
load_tensors: offloading 0 repeating layers to GPU
load_tensors: offloaded 0/29 layers to GPU
load_tensors:          CPU model buffer size =   166.92 MiB
load_tensors:    CUDA_Host model buffer size =  1123.71 MiB
llama_context: constructing llama_context
llama_context: n_seq_max     = 2
llama_context: n_ctx         = 8192
llama_context: n_ctx_per_seq = 4096
llama_context: n_batch       = 1024
llama_context: n_ubatch      = 512
llama_context: causal_attn   = 1
llama_context: flash_attn    = 0
llama_context: freq_base     = 1000000.0
llama_context: freq_scale    = 1
llama_context: n_ctx_per_seq (4096) < n_ctx_train (40960) -- the full capacity of the model will not be utilized
llama_context:        CPU  output buffer size =     1.17 MiB
llama_kv_cache_unified: kv_size = 8192, type_k = 'f16', type_v = 'f16', n_layer = 28, can_shift = 1, padding = 32
llama_kv_cache_unified:        CPU KV buffer size =   896.00 MiB
llama_kv_cache_unified: KV self size  =  896.00 MiB, K (f16):  448.00 MiB, V (f16):  448.00 MiB
llama_context:      CUDA0 compute buffer size =   570.93 MiB
llama_context:  CUDA_Host compute buffer size =    20.01 MiB
llama_context: graph nodes  = 1070
llama_context: graph splits = 368 (with bs=512), 1 (with bs=1)
time=2025-07-15T09:46:12.570+08:00 level=INFO source=server.go:637 msg="llama runner started in 1.25 seconds"
[GIN] 2025/07/15 - 09:46:12 | 200 |    1.5786592s |       127.0.0.1 | POST     "/api/generate"
C:\a\ollama\ollama\ml\backend\ggml\ggml\src\ggml-cuda\ggml-cuda.cu:2641: GGML_ASSERT(stat == cudaSuccess) failed
time=2025-07-15T09:46:19.445+08:00 level=ERROR source=server.go:807 msg="post predict" error="Post \"http://127.0.0.1:59254/completion\": read tcp 127.0.0.1:59256->127.0.0.1:59254: wsarecv: An existing connection was forcibly closed by the remote host."
[GIN] 2025/07/15 - 09:46:19 | 200 |    293.5268ms |       127.0.0.1 | POST     "/api/chat"
time=2025-07-15T09:46:19.563+08:00 level=ERROR source=server.go:464 msg="llama runner terminated" error="exit status 0xc0000409"

OS

Windows

GPU

AMD

CPU

AMD

Ollama version

0.9.6

Originally created by @tts2 on GitHub (Jul 15, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/11421 ### What is the issue? Windows 11 24H2 AMD Ryzen 7 PRO 8845HS w/ Radeon 780M Graphics 3.80 GHz 32.0 GB ### Relevant log output ```shell time=2025-07-15T09:31:26.800+08:00 level=INFO source=routes.go:1235 msg="server config" env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:4096 OLLAMA_DEBUG:INFO OLLAMA_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:C:\\Users\\tts2\\.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://* vscode-file://*] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES:]" time=2025-07-15T09:31:26.810+08:00 level=INFO source=images.go:476 msg="total blobs: 8" time=2025-07-15T09:31:26.811+08:00 level=INFO source=images.go:483 msg="total unused blobs removed: 0" time=2025-07-15T09:31:26.811+08:00 level=INFO source=routes.go:1288 msg="Listening on [::]:11434 (version 0.9.6)" time=2025-07-15T09:31:26.811+08:00 level=INFO source=gpu.go:217 msg="looking for compatible GPUs" time=2025-07-15T09:31:26.811+08:00 level=INFO source=gpu_windows.go:167 msg=packages count=1 time=2025-07-15T09:31:26.811+08:00 level=INFO source=gpu_windows.go:214 msg="" package=0 cores=8 efficiency=0 threads=16 time=2025-07-15T09:31:26.878+08:00 level=INFO source=gpu.go:272 msg="error looking up nvidia GPU memory" error="cuda driver library failed to get device context 801" time=2025-07-15T09:31:27.217+08:00 level=WARN source=amd_windows.go:138 msg="amdgpu is not supported (supported types:[gfx1030 gfx1100 gfx1101 gfx1102 gfx1151 gfx906])" gpu_type=gfx1103 gpu=0 library=C:\Users\tts2\AppData\Local\Programs\Ollama\lib\ollama\rocm time=2025-07-15T09:31:27.217+08:00 level=INFO source=gpu.go:377 msg="no compatible GPUs were discovered" time=2025-07-15T09:31:27.220+08:00 level=INFO source=types.go:130 msg="inference compute" id=0 library=cpu variant="" compute="" driver=0.0 name="" total="31.3 GiB" available="20.9 GiB" time=2025-07-15T09:31:37.210+08:00 level=INFO source=server.go:135 msg="system memory" total="31.3 GiB" free="20.7 GiB" free_swap="17.4 GiB" time=2025-07-15T09:31:37.211+08:00 level=INFO source=server.go:175 msg=offload library=cpu layers.requested=-1 layers.model=29 layers.offload=0 layers.split="" memory.available="[20.7 GiB]" memory.gpu_overhead="0 B" memory.required.full="2.3 GiB" memory.required.partial="0 B" memory.required.kv="896.0 MiB" memory.required.allocations="[2.3 GiB]" memory.weights.total="1.1 GiB" memory.weights.repeating="880.3 MiB" memory.weights.nonrepeating="243.4 MiB" memory.graph.full="298.7 MiB" memory.graph.partial="298.7 MiB" llama_model_loader: loaded meta data with 27 key-value pairs and 311 tensors from C:\Users\tts2\.ollama\models\blobs\sha256-3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6 (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 = qwen3 llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.name str = Qwen3 1.7B llama_model_loader: - kv 3: general.basename str = Qwen3 llama_model_loader: - kv 4: general.size_label str = 1.7B llama_model_loader: - kv 5: qwen3.block_count u32 = 28 llama_model_loader: - kv 6: qwen3.context_length u32 = 40960 llama_model_loader: - kv 7: qwen3.embedding_length u32 = 2048 llama_model_loader: - kv 8: qwen3.feed_forward_length u32 = 6144 llama_model_loader: - kv 9: qwen3.attention.head_count u32 = 16 llama_model_loader: - kv 10: qwen3.attention.head_count_kv u32 = 8 llama_model_loader: - kv 11: qwen3.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 12: qwen3.attention.layer_norm_rms_epsilon f32 = 0.000001 llama_model_loader: - kv 13: qwen3.attention.key_length u32 = 128 llama_model_loader: - kv 14: qwen3.attention.value_length u32 = 128 llama_model_loader: - kv 15: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 16: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 17: tokenizer.ggml.tokens arr[str,151936] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 18: tokenizer.ggml.token_type arr[i32,151936] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 19: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151645 llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 22: tokenizer.ggml.bos_token_id u32 = 151643 llama_model_loader: - kv 23: tokenizer.ggml.add_bos_token bool = false llama_model_loader: - kv 24: tokenizer.chat_template str = {%- if tools %}\n {{- '<|im_start|>... llama_model_loader: - kv 25: general.quantization_version u32 = 2 llama_model_loader: - kv 26: general.file_type u32 = 15 llama_model_loader: - type f32: 113 tensors llama_model_loader: - type f16: 28 tensors llama_model_loader: - type q4_K: 155 tensors llama_model_loader: - type q6_K: 15 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 1.26 GiB (5.33 BPW) load: special tokens cache size = 26 load: token to piece cache size = 0.9311 MB print_info: arch = qwen3 print_info: vocab_only = 1 print_info: model type = ?B print_info: model params = 2.03 B print_info: general.name = Qwen3 1.7B print_info: vocab type = BPE print_info: n_vocab = 151936 print_info: n_merges = 151387 print_info: BOS token = 151643 '<|endoftext|>' print_info: EOS token = 151645 '<|im_end|>' print_info: EOT token = 151645 '<|im_end|>' print_info: PAD token = 151643 '<|endoftext|>' print_info: LF token = 198 'Ċ' print_info: FIM PRE token = 151659 '<|fim_prefix|>' print_info: FIM SUF token = 151661 '<|fim_suffix|>' print_info: FIM MID token = 151660 '<|fim_middle|>' print_info: FIM PAD token = 151662 '<|fim_pad|>' print_info: FIM REP token = 151663 '<|repo_name|>' print_info: FIM SEP token = 151664 '<|file_sep|>' print_info: EOG token = 151643 '<|endoftext|>' print_info: EOG token = 151645 '<|im_end|>' print_info: EOG token = 151662 '<|fim_pad|>' print_info: EOG token = 151663 '<|repo_name|>' print_info: EOG token = 151664 '<|file_sep|>' print_info: max token length = 256 llama_model_load: vocab only - skipping tensors time=2025-07-15T09:31:37.430+08:00 level=INFO source=server.go:438 msg="starting llama server" cmd="C:\\Users\\tts2\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --model C:\\Users\\tts2\\.ollama\\models\\blobs\\sha256-3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6 --ctx-size 8192 --batch-size 512 --threads 8 --no-mmap --parallel 2 --port 59111" time=2025-07-15T09:31:37.442+08:00 level=INFO source=sched.go:483 msg="loaded runners" count=1 time=2025-07-15T09:31:37.442+08:00 level=INFO source=server.go:598 msg="waiting for llama runner to start responding" time=2025-07-15T09:31:37.442+08:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server error" time=2025-07-15T09:31:37.476+08:00 level=INFO source=runner.go:815 msg="starting go runner" ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no ggml_cuda_init: found 1 CUDA devices: Device 0: AMD Radeon 780M Graphics [ZLUDA], compute capability 8.8, VMM: no load_backend: loaded CUDA backend from C:\Users\tts2\AppData\Local\Programs\Ollama\lib\ollama\ggml-cuda.dll load_backend: loaded CPU backend from C:\Users\tts2\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-icelake.dll time=2025-07-15T09:31:37.585+08:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.AVX512=1 CPU.0.AVX512_VBMI=1 CPU.0.AVX512_VNNI=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 CUDA.0.ARCHS=500,600,610,700,750,800,860,870,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(clang) time=2025-07-15T09:31:37.586+08:00 level=INFO source=runner.go:874 msg="Server listening on 127.0.0.1:59111" time=2025-07-15T09:31:37.693+08:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server loading model" llama_model_load_from_file_impl: using device CUDA0 (AMD Radeon 780M Graphics [ZLUDA]) - 12391 MiB free llama_model_loader: loaded meta data with 27 key-value pairs and 311 tensors from C:\Users\tts2\.ollama\models\blobs\sha256-3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6 (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 = qwen3 llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.name str = Qwen3 1.7B llama_model_loader: - kv 3: general.basename str = Qwen3 llama_model_loader: - kv 4: general.size_label str = 1.7B llama_model_loader: - kv 5: qwen3.block_count u32 = 28 llama_model_loader: - kv 6: qwen3.context_length u32 = 40960 llama_model_loader: - kv 7: qwen3.embedding_length u32 = 2048 llama_model_loader: - kv 8: qwen3.feed_forward_length u32 = 6144 llama_model_loader: - kv 9: qwen3.attention.head_count u32 = 16 llama_model_loader: - kv 10: qwen3.attention.head_count_kv u32 = 8 llama_model_loader: - kv 11: qwen3.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 12: qwen3.attention.layer_norm_rms_epsilon f32 = 0.000001 llama_model_loader: - kv 13: qwen3.attention.key_length u32 = 128 llama_model_loader: - kv 14: qwen3.attention.value_length u32 = 128 llama_model_loader: - kv 15: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 16: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 17: tokenizer.ggml.tokens arr[str,151936] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 18: tokenizer.ggml.token_type arr[i32,151936] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 19: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151645 llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 22: tokenizer.ggml.bos_token_id u32 = 151643 llama_model_loader: - kv 23: tokenizer.ggml.add_bos_token bool = false llama_model_loader: - kv 24: tokenizer.chat_template str = {%- if tools %}\n {{- '<|im_start|>... llama_model_loader: - kv 25: general.quantization_version u32 = 2 llama_model_loader: - kv 26: general.file_type u32 = 15 llama_model_loader: - type f32: 113 tensors llama_model_loader: - type f16: 28 tensors llama_model_loader: - type q4_K: 155 tensors llama_model_loader: - type q6_K: 15 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 1.26 GiB (5.33 BPW) load: special tokens cache size = 26 load: token to piece cache size = 0.9311 MB print_info: arch = qwen3 print_info: vocab_only = 0 print_info: n_ctx_train = 40960 print_info: n_embd = 2048 print_info: n_layer = 28 print_info: n_head = 16 print_info: n_head_kv = 8 print_info: n_rot = 128 print_info: n_swa = 0 print_info: n_swa_pattern = 1 print_info: n_embd_head_k = 128 print_info: n_embd_head_v = 128 print_info: n_gqa = 2 print_info: n_embd_k_gqa = 1024 print_info: n_embd_v_gqa = 1024 print_info: f_norm_eps = 0.0e+00 print_info: f_norm_rms_eps = 1.0e-06 print_info: f_clamp_kqv = 0.0e+00 print_info: f_max_alibi_bias = 0.0e+00 print_info: f_logit_scale = 0.0e+00 print_info: f_attn_scale = 0.0e+00 print_info: n_ff = 6144 print_info: n_expert = 0 print_info: n_expert_used = 0 print_info: causal attn = 1 print_info: pooling type = 0 print_info: rope type = 2 print_info: rope scaling = linear print_info: freq_base_train = 1000000.0 print_info: freq_scale_train = 1 print_info: n_ctx_orig_yarn = 40960 print_info: rope_finetuned = unknown print_info: ssm_d_conv = 0 print_info: ssm_d_inner = 0 print_info: ssm_d_state = 0 print_info: ssm_dt_rank = 0 print_info: ssm_dt_b_c_rms = 0 print_info: model type = 1.7B print_info: model params = 2.03 B print_info: general.name = Qwen3 1.7B print_info: vocab type = BPE print_info: n_vocab = 151936 print_info: n_merges = 151387 print_info: BOS token = 151643 '<|endoftext|>' print_info: EOS token = 151645 '<|im_end|>' print_info: EOT token = 151645 '<|im_end|>' print_info: PAD token = 151643 '<|endoftext|>' print_info: LF token = 198 'Ċ' print_info: FIM PRE token = 151659 '<|fim_prefix|>' print_info: FIM SUF token = 151661 '<|fim_suffix|>' print_info: FIM MID token = 151660 '<|fim_middle|>' print_info: FIM PAD token = 151662 '<|fim_pad|>' print_info: FIM REP token = 151663 '<|repo_name|>' print_info: FIM SEP token = 151664 '<|file_sep|>' print_info: EOG token = 151643 '<|endoftext|>' print_info: EOG token = 151645 '<|im_end|>' print_info: EOG token = 151662 '<|fim_pad|>' print_info: EOG token = 151663 '<|repo_name|>' print_info: EOG token = 151664 '<|file_sep|>' print_info: max token length = 256 load_tensors: loading model tensors, this can take a while... (mmap = false) load_tensors: offloading 0 repeating layers to GPU load_tensors: offloaded 0/29 layers to GPU load_tensors: CPU model buffer size = 166.92 MiB load_tensors: CUDA_Host model buffer size = 1123.71 MiB llama_context: constructing llama_context llama_context: n_seq_max = 2 llama_context: n_ctx = 8192 llama_context: n_ctx_per_seq = 4096 llama_context: n_batch = 1024 llama_context: n_ubatch = 512 llama_context: causal_attn = 1 llama_context: flash_attn = 0 llama_context: freq_base = 1000000.0 llama_context: freq_scale = 1 llama_context: n_ctx_per_seq (4096) < n_ctx_train (40960) -- the full capacity of the model will not be utilized llama_context: CPU output buffer size = 1.17 MiB llama_kv_cache_unified: kv_size = 8192, type_k = 'f16', type_v = 'f16', n_layer = 28, can_shift = 1, padding = 32 llama_kv_cache_unified: CPU KV buffer size = 896.00 MiB llama_kv_cache_unified: KV self size = 896.00 MiB, K (f16): 448.00 MiB, V (f16): 448.00 MiB llama_context: CUDA0 compute buffer size = 570.93 MiB llama_context: CUDA_Host compute buffer size = 20.01 MiB llama_context: graph nodes = 1070 llama_context: graph splits = 368 (with bs=512), 1 (with bs=1) time=2025-07-15T09:31:38.695+08:00 level=INFO source=server.go:637 msg="llama runner started in 1.25 seconds" C:\a\ollama\ollama\ml\backend\ggml\ggml\src\ggml-cuda\ggml-cuda.cu:2641: GGML_ASSERT(stat == cudaSuccess) failed time=2025-07-15T09:31:38.954+08:00 level=ERROR source=server.go:807 msg="post predict" error="Post \"http://127.0.0.1:59111/completion\": read tcp 127.0.0.1:59113->127.0.0.1:59111: wsarecv: An existing connection was forcibly closed by the remote host." [GIN] 2025/07/15 - 09:31:38 | 500 | 1.8302656s | 192.168.149.120 | POST "/v1/chat/completions" time=2025-07-15T09:31:39.066+08:00 level=ERROR source=server.go:464 msg="llama runner terminated" error="exit status 0xc0000409" time=2025-07-15T09:31:39.503+08:00 level=INFO source=server.go:135 msg="system memory" total="31.3 GiB" free="20.6 GiB" free_swap="17.3 GiB" time=2025-07-15T09:31:39.503+08:00 level=INFO source=server.go:175 msg=offload library=cpu layers.requested=-1 layers.model=29 layers.offload=0 layers.split="" memory.available="[20.6 GiB]" memory.gpu_overhead="0 B" memory.required.full="2.3 GiB" memory.required.partial="0 B" memory.required.kv="896.0 MiB" memory.required.allocations="[2.3 GiB]" memory.weights.total="1.1 GiB" memory.weights.repeating="880.3 MiB" memory.weights.nonrepeating="243.4 MiB" memory.graph.full="298.7 MiB" memory.graph.partial="298.7 MiB" llama_model_loader: loaded meta data with 27 key-value pairs and 311 tensors from C:\Users\tts2\.ollama\models\blobs\sha256-3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6 (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 = qwen3 llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.name str = Qwen3 1.7B llama_model_loader: - kv 3: general.basename str = Qwen3 llama_model_loader: - kv 4: general.size_label str = 1.7B llama_model_loader: - kv 5: qwen3.block_count u32 = 28 llama_model_loader: - kv 6: qwen3.context_length u32 = 40960 llama_model_loader: - kv 7: qwen3.embedding_length u32 = 2048 llama_model_loader: - kv 8: qwen3.feed_forward_length u32 = 6144 llama_model_loader: - kv 9: qwen3.attention.head_count u32 = 16 llama_model_loader: - kv 10: qwen3.attention.head_count_kv u32 = 8 llama_model_loader: - kv 11: qwen3.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 12: qwen3.attention.layer_norm_rms_epsilon f32 = 0.000001 llama_model_loader: - kv 13: qwen3.attention.key_length u32 = 128 llama_model_loader: - kv 14: qwen3.attention.value_length u32 = 128 llama_model_loader: - kv 15: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 16: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 17: tokenizer.ggml.tokens arr[str,151936] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 18: tokenizer.ggml.token_type arr[i32,151936] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 19: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151645 llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 22: tokenizer.ggml.bos_token_id u32 = 151643 llama_model_loader: - kv 23: tokenizer.ggml.add_bos_token bool = false llama_model_loader: - kv 24: tokenizer.chat_template str = {%- if tools %}\n {{- '<|im_start|>... llama_model_loader: - kv 25: general.quantization_version u32 = 2 llama_model_loader: - kv 26: general.file_type u32 = 15 llama_model_loader: - type f32: 113 tensors llama_model_loader: - type f16: 28 tensors llama_model_loader: - type q4_K: 155 tensors llama_model_loader: - type q6_K: 15 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 1.26 GiB (5.33 BPW) load: special tokens cache size = 26 load: token to piece cache size = 0.9311 MB print_info: arch = qwen3 print_info: vocab_only = 1 print_info: model type = ?B print_info: model params = 2.03 B print_info: general.name = Qwen3 1.7B print_info: vocab type = BPE print_info: n_vocab = 151936 print_info: n_merges = 151387 print_info: BOS token = 151643 '<|endoftext|>' print_info: EOS token = 151645 '<|im_end|>' print_info: EOT token = 151645 '<|im_end|>' print_info: PAD token = 151643 '<|endoftext|>' print_info: LF token = 198 'Ċ' print_info: FIM PRE token = 151659 '<|fim_prefix|>' print_info: FIM SUF token = 151661 '<|fim_suffix|>' print_info: FIM MID token = 151660 '<|fim_middle|>' print_info: FIM PAD token = 151662 '<|fim_pad|>' print_info: FIM REP token = 151663 '<|repo_name|>' print_info: FIM SEP token = 151664 '<|file_sep|>' print_info: EOG token = 151643 '<|endoftext|>' print_info: EOG token = 151645 '<|im_end|>' print_info: EOG token = 151662 '<|fim_pad|>' print_info: EOG token = 151663 '<|repo_name|>' print_info: EOG token = 151664 '<|file_sep|>' print_info: max token length = 256 llama_model_load: vocab only - skipping tensors time=2025-07-15T09:31:39.726+08:00 level=INFO source=server.go:438 msg="starting llama server" cmd="C:\\Users\\tts2\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --model C:\\Users\\tts2\\.ollama\\models\\blobs\\sha256-3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6 --ctx-size 8192 --batch-size 512 --threads 8 --no-mmap --parallel 2 --port 59114" time=2025-07-15T09:31:39.738+08:00 level=INFO source=sched.go:483 msg="loaded runners" count=1 time=2025-07-15T09:31:39.738+08:00 level=INFO source=server.go:598 msg="waiting for llama runner to start responding" time=2025-07-15T09:31:39.738+08:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server error" time=2025-07-15T09:31:39.772+08:00 level=INFO source=runner.go:815 msg="starting go runner" ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no ggml_cuda_init: found 1 CUDA devices: Device 0: AMD Radeon 780M Graphics [ZLUDA], compute capability 8.8, VMM: no load_backend: loaded CUDA backend from C:\Users\tts2\AppData\Local\Programs\Ollama\lib\ollama\ggml-cuda.dll load_backend: loaded CPU backend from C:\Users\tts2\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-icelake.dll time=2025-07-15T09:31:39.882+08:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.AVX512=1 CPU.0.AVX512_VBMI=1 CPU.0.AVX512_VNNI=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 CUDA.0.ARCHS=500,600,610,700,750,800,860,870,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(clang) time=2025-07-15T09:31:39.883+08:00 level=INFO source=runner.go:874 msg="Server listening on 127.0.0.1:59114" time=2025-07-15T09:31:39.990+08:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server loading model" llama_model_load_from_file_impl: using device CUDA0 (AMD Radeon 780M Graphics [ZLUDA]) - 12391 MiB free llama_model_loader: loaded meta data with 27 key-value pairs and 311 tensors from C:\Users\tts2\.ollama\models\blobs\sha256-3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6 (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 = qwen3 llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.name str = Qwen3 1.7B llama_model_loader: - kv 3: general.basename str = Qwen3 llama_model_loader: - kv 4: general.size_label str = 1.7B llama_model_loader: - kv 5: qwen3.block_count u32 = 28 llama_model_loader: - kv 6: qwen3.context_length u32 = 40960 llama_model_loader: - kv 7: qwen3.embedding_length u32 = 2048 llama_model_loader: - kv 8: qwen3.feed_forward_length u32 = 6144 llama_model_loader: - kv 9: qwen3.attention.head_count u32 = 16 llama_model_loader: - kv 10: qwen3.attention.head_count_kv u32 = 8 llama_model_loader: - kv 11: qwen3.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 12: qwen3.attention.layer_norm_rms_epsilon f32 = 0.000001 llama_model_loader: - kv 13: qwen3.attention.key_length u32 = 128 llama_model_loader: - kv 14: qwen3.attention.value_length u32 = 128 llama_model_loader: - kv 15: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 16: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 17: tokenizer.ggml.tokens arr[str,151936] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 18: tokenizer.ggml.token_type arr[i32,151936] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 19: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151645 llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 22: tokenizer.ggml.bos_token_id u32 = 151643 llama_model_loader: - kv 23: tokenizer.ggml.add_bos_token bool = false llama_model_loader: - kv 24: tokenizer.chat_template str = {%- if tools %}\n {{- '<|im_start|>... llama_model_loader: - kv 25: general.quantization_version u32 = 2 llama_model_loader: - kv 26: general.file_type u32 = 15 llama_model_loader: - type f32: 113 tensors llama_model_loader: - type f16: 28 tensors llama_model_loader: - type q4_K: 155 tensors llama_model_loader: - type q6_K: 15 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 1.26 GiB (5.33 BPW) load: special tokens cache size = 26 load: token to piece cache size = 0.9311 MB print_info: arch = qwen3 print_info: vocab_only = 0 print_info: n_ctx_train = 40960 print_info: n_embd = 2048 print_info: n_layer = 28 print_info: n_head = 16 print_info: n_head_kv = 8 print_info: n_rot = 128 print_info: n_swa = 0 print_info: n_swa_pattern = 1 print_info: n_embd_head_k = 128 print_info: n_embd_head_v = 128 print_info: n_gqa = 2 print_info: n_embd_k_gqa = 1024 print_info: n_embd_v_gqa = 1024 print_info: f_norm_eps = 0.0e+00 print_info: f_norm_rms_eps = 1.0e-06 print_info: f_clamp_kqv = 0.0e+00 print_info: f_max_alibi_bias = 0.0e+00 print_info: f_logit_scale = 0.0e+00 print_info: f_attn_scale = 0.0e+00 print_info: n_ff = 6144 print_info: n_expert = 0 print_info: n_expert_used = 0 print_info: causal attn = 1 print_info: pooling type = 0 print_info: rope type = 2 print_info: rope scaling = linear print_info: freq_base_train = 1000000.0 print_info: freq_scale_train = 1 print_info: n_ctx_orig_yarn = 40960 print_info: rope_finetuned = unknown print_info: ssm_d_conv = 0 print_info: ssm_d_inner = 0 print_info: ssm_d_state = 0 print_info: ssm_dt_rank = 0 print_info: ssm_dt_b_c_rms = 0 print_info: model type = 1.7B print_info: model params = 2.03 B print_info: general.name = Qwen3 1.7B print_info: vocab type = BPE print_info: n_vocab = 151936 print_info: n_merges = 151387 print_info: BOS token = 151643 '<|endoftext|>' print_info: EOS token = 151645 '<|im_end|>' print_info: EOT token = 151645 '<|im_end|>' print_info: PAD token = 151643 '<|endoftext|>' print_info: LF token = 198 'Ċ' print_info: FIM PRE token = 151659 '<|fim_prefix|>' print_info: FIM SUF token = 151661 '<|fim_suffix|>' print_info: FIM MID token = 151660 '<|fim_middle|>' print_info: FIM PAD token = 151662 '<|fim_pad|>' print_info: FIM REP token = 151663 '<|repo_name|>' print_info: FIM SEP token = 151664 '<|file_sep|>' print_info: EOG token = 151643 '<|endoftext|>' print_info: EOG token = 151645 '<|im_end|>' print_info: EOG token = 151662 '<|fim_pad|>' print_info: EOG token = 151663 '<|repo_name|>' print_info: EOG token = 151664 '<|file_sep|>' print_info: max token length = 256 load_tensors: loading model tensors, this can take a while... (mmap = false) load_tensors: offloading 0 repeating layers to GPU load_tensors: offloaded 0/29 layers to GPU load_tensors: CPU model buffer size = 166.92 MiB load_tensors: CUDA_Host model buffer size = 1123.71 MiB llama_context: constructing llama_context llama_context: n_seq_max = 2 llama_context: n_ctx = 8192 llama_context: n_ctx_per_seq = 4096 llama_context: n_batch = 1024 llama_context: n_ubatch = 512 llama_context: causal_attn = 1 llama_context: flash_attn = 0 llama_context: freq_base = 1000000.0 llama_context: freq_scale = 1 llama_context: n_ctx_per_seq (4096) < n_ctx_train (40960) -- the full capacity of the model will not be utilized llama_context: CPU output buffer size = 1.17 MiB llama_kv_cache_unified: kv_size = 8192, type_k = 'f16', type_v = 'f16', n_layer = 28, can_shift = 1, padding = 32 llama_kv_cache_unified: CPU KV buffer size = 896.00 MiB llama_kv_cache_unified: KV self size = 896.00 MiB, K (f16): 448.00 MiB, V (f16): 448.00 MiB llama_context: CUDA0 compute buffer size = 570.93 MiB llama_context: CUDA_Host compute buffer size = 20.01 MiB llama_context: graph nodes = 1070 llama_context: graph splits = 368 (with bs=512), 1 (with bs=1) time=2025-07-15T09:31:40.991+08:00 level=INFO source=server.go:637 msg="llama runner started in 1.25 seconds" C:\a\ollama\ollama\ml\backend\ggml\ggml\src\ggml-cuda\ggml-cuda.cu:2641: GGML_ASSERT(stat == cudaSuccess) failed time=2025-07-15T09:31:41.197+08:00 level=ERROR source=server.go:807 msg="post predict" error="Post \"http://127.0.0.1:59114/completion\": read tcp 127.0.0.1:59116->127.0.0.1:59114: wsarecv: An existing connection was forcibly closed by the remote host." [GIN] 2025/07/15 - 09:31:41 | 500 | 1.8284561s | 192.168.149.120 | POST "/v1/chat/completions" time=2025-07-15T09:31:41.304+08:00 level=ERROR source=server.go:464 msg="llama runner terminated" error="exit status 0xc0000409" time=2025-07-15T09:31:42.173+08:00 level=INFO source=server.go:135 msg="system memory" total="31.3 GiB" free="20.6 GiB" free_swap="17.3 GiB" time=2025-07-15T09:31:42.173+08:00 level=INFO source=server.go:175 msg=offload library=cpu layers.requested=-1 layers.model=29 layers.offload=0 layers.split="" memory.available="[20.6 GiB]" memory.gpu_overhead="0 B" memory.required.full="2.3 GiB" memory.required.partial="0 B" memory.required.kv="896.0 MiB" memory.required.allocations="[2.3 GiB]" memory.weights.total="1.1 GiB" memory.weights.repeating="880.3 MiB" memory.weights.nonrepeating="243.4 MiB" memory.graph.full="298.7 MiB" memory.graph.partial="298.7 MiB" llama_model_loader: loaded meta data with 27 key-value pairs and 311 tensors from C:\Users\tts2\.ollama\models\blobs\sha256-3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6 (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 = qwen3 llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.name str = Qwen3 1.7B llama_model_loader: - kv 3: general.basename str = Qwen3 llama_model_loader: - kv 4: general.size_label str = 1.7B llama_model_loader: - kv 5: qwen3.block_count u32 = 28 llama_model_loader: - kv 6: qwen3.context_length u32 = 40960 llama_model_loader: - kv 7: qwen3.embedding_length u32 = 2048 llama_model_loader: - kv 8: qwen3.feed_forward_length u32 = 6144 llama_model_loader: - kv 9: qwen3.attention.head_count u32 = 16 llama_model_loader: - kv 10: qwen3.attention.head_count_kv u32 = 8 llama_model_loader: - kv 11: qwen3.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 12: qwen3.attention.layer_norm_rms_epsilon f32 = 0.000001 llama_model_loader: - kv 13: qwen3.attention.key_length u32 = 128 llama_model_loader: - kv 14: qwen3.attention.value_length u32 = 128 llama_model_loader: - kv 15: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 16: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 17: tokenizer.ggml.tokens arr[str,151936] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 18: tokenizer.ggml.token_type arr[i32,151936] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 19: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151645 llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 22: tokenizer.ggml.bos_token_id u32 = 151643 llama_model_loader: - kv 23: tokenizer.ggml.add_bos_token bool = false llama_model_loader: - kv 24: tokenizer.chat_template str = {%- if tools %}\n {{- '<|im_start|>... llama_model_loader: - kv 25: general.quantization_version u32 = 2 llama_model_loader: - kv 26: general.file_type u32 = 15 llama_model_loader: - type f32: 113 tensors llama_model_loader: - type f16: 28 tensors llama_model_loader: - type q4_K: 155 tensors llama_model_loader: - type q6_K: 15 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 1.26 GiB (5.33 BPW) load: special tokens cache size = 26 load: token to piece cache size = 0.9311 MB print_info: arch = qwen3 print_info: vocab_only = 1 print_info: model type = ?B print_info: model params = 2.03 B print_info: general.name = Qwen3 1.7B print_info: vocab type = BPE print_info: n_vocab = 151936 print_info: n_merges = 151387 print_info: BOS token = 151643 '<|endoftext|>' print_info: EOS token = 151645 '<|im_end|>' print_info: EOT token = 151645 '<|im_end|>' print_info: PAD token = 151643 '<|endoftext|>' print_info: LF token = 198 'Ċ' print_info: FIM PRE token = 151659 '<|fim_prefix|>' print_info: FIM SUF token = 151661 '<|fim_suffix|>' print_info: FIM MID token = 151660 '<|fim_middle|>' print_info: FIM PAD token = 151662 '<|fim_pad|>' print_info: FIM REP token = 151663 '<|repo_name|>' print_info: FIM SEP token = 151664 '<|file_sep|>' print_info: EOG token = 151643 '<|endoftext|>' print_info: EOG token = 151645 '<|im_end|>' print_info: EOG token = 151662 '<|fim_pad|>' print_info: EOG token = 151663 '<|repo_name|>' print_info: EOG token = 151664 '<|file_sep|>' print_info: max token length = 256 llama_model_load: vocab only - skipping tensors time=2025-07-15T09:31:42.374+08:00 level=INFO source=server.go:438 msg="starting llama server" cmd="C:\\Users\\tts2\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --model C:\\Users\\tts2\\.ollama\\models\\blobs\\sha256-3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6 --ctx-size 8192 --batch-size 512 --threads 8 --no-mmap --parallel 2 --port 59117" time=2025-07-15T09:31:42.386+08:00 level=INFO source=sched.go:483 msg="loaded runners" count=1 time=2025-07-15T09:31:42.386+08:00 level=INFO source=server.go:598 msg="waiting for llama runner to start responding" time=2025-07-15T09:31:42.386+08:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server error" time=2025-07-15T09:31:42.419+08:00 level=INFO source=runner.go:815 msg="starting go runner" ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no ggml_cuda_init: found 1 CUDA devices: Device 0: AMD Radeon 780M Graphics [ZLUDA], compute capability 8.8, VMM: no load_backend: loaded CUDA backend from C:\Users\tts2\AppData\Local\Programs\Ollama\lib\ollama\ggml-cuda.dll load_backend: loaded CPU backend from C:\Users\tts2\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-icelake.dll time=2025-07-15T09:31:42.530+08:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.AVX512=1 CPU.0.AVX512_VBMI=1 CPU.0.AVX512_VNNI=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 CUDA.0.ARCHS=500,600,610,700,750,800,860,870,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(clang) time=2025-07-15T09:31:42.530+08:00 level=INFO source=runner.go:874 msg="Server listening on 127.0.0.1:59117" time=2025-07-15T09:31:42.637+08:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server loading model" llama_model_load_from_file_impl: using device CUDA0 (AMD Radeon 780M Graphics [ZLUDA]) - 12391 MiB free llama_model_loader: loaded meta data with 27 key-value pairs and 311 tensors from C:\Users\tts2\.ollama\models\blobs\sha256-3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6 (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 = qwen3 llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.name str = Qwen3 1.7B llama_model_loader: - kv 3: general.basename str = Qwen3 llama_model_loader: - kv 4: general.size_label str = 1.7B llama_model_loader: - kv 5: qwen3.block_count u32 = 28 llama_model_loader: - kv 6: qwen3.context_length u32 = 40960 llama_model_loader: - kv 7: qwen3.embedding_length u32 = 2048 llama_model_loader: - kv 8: qwen3.feed_forward_length u32 = 6144 llama_model_loader: - kv 9: qwen3.attention.head_count u32 = 16 llama_model_loader: - kv 10: qwen3.attention.head_count_kv u32 = 8 llama_model_loader: - kv 11: qwen3.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 12: qwen3.attention.layer_norm_rms_epsilon f32 = 0.000001 llama_model_loader: - kv 13: qwen3.attention.key_length u32 = 128 llama_model_loader: - kv 14: qwen3.attention.value_length u32 = 128 llama_model_loader: - kv 15: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 16: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 17: tokenizer.ggml.tokens arr[str,151936] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 18: tokenizer.ggml.token_type arr[i32,151936] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 19: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151645 llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 22: tokenizer.ggml.bos_token_id u32 = 151643 llama_model_loader: - kv 23: tokenizer.ggml.add_bos_token bool = false llama_model_loader: - kv 24: tokenizer.chat_template str = {%- if tools %}\n {{- '<|im_start|>... llama_model_loader: - kv 25: general.quantization_version u32 = 2 llama_model_loader: - kv 26: general.file_type u32 = 15 llama_model_loader: - type f32: 113 tensors llama_model_loader: - type f16: 28 tensors llama_model_loader: - type q4_K: 155 tensors llama_model_loader: - type q6_K: 15 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 1.26 GiB (5.33 BPW) load: special tokens cache size = 26 load: token to piece cache size = 0.9311 MB print_info: arch = qwen3 print_info: vocab_only = 0 print_info: n_ctx_train = 40960 print_info: n_embd = 2048 print_info: n_layer = 28 print_info: n_head = 16 print_info: n_head_kv = 8 print_info: n_rot = 128 print_info: n_swa = 0 print_info: n_swa_pattern = 1 print_info: n_embd_head_k = 128 print_info: n_embd_head_v = 128 print_info: n_gqa = 2 print_info: n_embd_k_gqa = 1024 print_info: n_embd_v_gqa = 1024 print_info: f_norm_eps = 0.0e+00 print_info: f_norm_rms_eps = 1.0e-06 print_info: f_clamp_kqv = 0.0e+00 print_info: f_max_alibi_bias = 0.0e+00 print_info: f_logit_scale = 0.0e+00 print_info: f_attn_scale = 0.0e+00 print_info: n_ff = 6144 print_info: n_expert = 0 print_info: n_expert_used = 0 print_info: causal attn = 1 print_info: pooling type = 0 print_info: rope type = 2 print_info: rope scaling = linear print_info: freq_base_train = 1000000.0 print_info: freq_scale_train = 1 print_info: n_ctx_orig_yarn = 40960 print_info: rope_finetuned = unknown print_info: ssm_d_conv = 0 print_info: ssm_d_inner = 0 print_info: ssm_d_state = 0 print_info: ssm_dt_rank = 0 print_info: ssm_dt_b_c_rms = 0 print_info: model type = 1.7B print_info: model params = 2.03 B print_info: general.name = Qwen3 1.7B print_info: vocab type = BPE print_info: n_vocab = 151936 print_info: n_merges = 151387 print_info: BOS token = 151643 '<|endoftext|>' print_info: EOS token = 151645 '<|im_end|>' print_info: EOT token = 151645 '<|im_end|>' print_info: PAD token = 151643 '<|endoftext|>' print_info: LF token = 198 'Ċ' print_info: FIM PRE token = 151659 '<|fim_prefix|>' print_info: FIM SUF token = 151661 '<|fim_suffix|>' print_info: FIM MID token = 151660 '<|fim_middle|>' print_info: FIM PAD token = 151662 '<|fim_pad|>' print_info: FIM REP token = 151663 '<|repo_name|>' print_info: FIM SEP token = 151664 '<|file_sep|>' print_info: EOG token = 151643 '<|endoftext|>' print_info: EOG token = 151645 '<|im_end|>' print_info: EOG token = 151662 '<|fim_pad|>' print_info: EOG token = 151663 '<|repo_name|>' print_info: EOG token = 151664 '<|file_sep|>' print_info: max token length = 256 load_tensors: loading model tensors, this can take a while... (mmap = false) load_tensors: offloading 0 repeating layers to GPU load_tensors: offloaded 0/29 layers to GPU load_tensors: CPU model buffer size = 166.92 MiB load_tensors: CUDA_Host model buffer size = 1123.71 MiB llama_context: constructing llama_context llama_context: n_seq_max = 2 llama_context: n_ctx = 8192 llama_context: n_ctx_per_seq = 4096 llama_context: n_batch = 1024 llama_context: n_ubatch = 512 llama_context: causal_attn = 1 llama_context: flash_attn = 0 llama_context: freq_base = 1000000.0 llama_context: freq_scale = 1 llama_context: n_ctx_per_seq (4096) < n_ctx_train (40960) -- the full capacity of the model will not be utilized llama_context: CPU output buffer size = 1.17 MiB llama_kv_cache_unified: kv_size = 8192, type_k = 'f16', type_v = 'f16', n_layer = 28, can_shift = 1, padding = 32 llama_kv_cache_unified: CPU KV buffer size = 896.00 MiB llama_kv_cache_unified: KV self size = 896.00 MiB, K (f16): 448.00 MiB, V (f16): 448.00 MiB llama_context: CUDA0 compute buffer size = 570.93 MiB llama_context: CUDA_Host compute buffer size = 20.01 MiB llama_context: graph nodes = 1070 llama_context: graph splits = 368 (with bs=512), 1 (with bs=1) time=2025-07-15T09:31:43.639+08:00 level=INFO source=server.go:637 msg="llama runner started in 1.25 seconds" C:\a\ollama\ollama\ml\backend\ggml\ggml\src\ggml-cuda\ggml-cuda.cu:2641: GGML_ASSERT(stat == cudaSuccess) failed time=2025-07-15T09:31:43.838+08:00 level=ERROR source=server.go:807 msg="post predict" error="Post \"http://127.0.0.1:59117/completion\": read tcp 127.0.0.1:59119->127.0.0.1:59117: wsarecv: An existing connection was forcibly closed by the remote host." [GIN] 2025/07/15 - 09:31:43 | 500 | 1.7839638s | 192.168.149.120 | POST "/v1/chat/completions" time=2025-07-15T09:31:43.938+08:00 level=ERROR source=server.go:464 msg="llama runner terminated" error="exit status 0xc0000409" [GIN] 2025/07/15 - 09:46:10 | 200 | 578.8µs | 127.0.0.1 | HEAD "/" [GIN] 2025/07/15 - 09:46:10 | 200 | 55.609ms | 127.0.0.1 | POST "/api/show" time=2025-07-15T09:46:11.074+08:00 level=INFO source=server.go:135 msg="system memory" total="31.3 GiB" free="20.5 GiB" free_swap="17.4 GiB" time=2025-07-15T09:46:11.074+08:00 level=INFO source=server.go:175 msg=offload library=cpu layers.requested=-1 layers.model=29 layers.offload=0 layers.split="" memory.available="[20.5 GiB]" memory.gpu_overhead="0 B" memory.required.full="2.3 GiB" memory.required.partial="0 B" memory.required.kv="896.0 MiB" memory.required.allocations="[2.3 GiB]" memory.weights.total="1.1 GiB" memory.weights.repeating="880.3 MiB" memory.weights.nonrepeating="243.4 MiB" memory.graph.full="298.7 MiB" memory.graph.partial="298.7 MiB" llama_model_loader: loaded meta data with 27 key-value pairs and 311 tensors from C:\Users\tts2\.ollama\models\blobs\sha256-3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6 (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 = qwen3 llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.name str = Qwen3 1.7B llama_model_loader: - kv 3: general.basename str = Qwen3 llama_model_loader: - kv 4: general.size_label str = 1.7B llama_model_loader: - kv 5: qwen3.block_count u32 = 28 llama_model_loader: - kv 6: qwen3.context_length u32 = 40960 llama_model_loader: - kv 7: qwen3.embedding_length u32 = 2048 llama_model_loader: - kv 8: qwen3.feed_forward_length u32 = 6144 llama_model_loader: - kv 9: qwen3.attention.head_count u32 = 16 llama_model_loader: - kv 10: qwen3.attention.head_count_kv u32 = 8 llama_model_loader: - kv 11: qwen3.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 12: qwen3.attention.layer_norm_rms_epsilon f32 = 0.000001 llama_model_loader: - kv 13: qwen3.attention.key_length u32 = 128 llama_model_loader: - kv 14: qwen3.attention.value_length u32 = 128 llama_model_loader: - kv 15: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 16: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 17: tokenizer.ggml.tokens arr[str,151936] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 18: tokenizer.ggml.token_type arr[i32,151936] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 19: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151645 llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 22: tokenizer.ggml.bos_token_id u32 = 151643 llama_model_loader: - kv 23: tokenizer.ggml.add_bos_token bool = false llama_model_loader: - kv 24: tokenizer.chat_template str = {%- if tools %}\n {{- '<|im_start|>... llama_model_loader: - kv 25: general.quantization_version u32 = 2 llama_model_loader: - kv 26: general.file_type u32 = 15 llama_model_loader: - type f32: 113 tensors llama_model_loader: - type f16: 28 tensors llama_model_loader: - type q4_K: 155 tensors llama_model_loader: - type q6_K: 15 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 1.26 GiB (5.33 BPW) load: special tokens cache size = 26 load: token to piece cache size = 0.9311 MB print_info: arch = qwen3 print_info: vocab_only = 1 print_info: model type = ?B print_info: model params = 2.03 B print_info: general.name = Qwen3 1.7B print_info: vocab type = BPE print_info: n_vocab = 151936 print_info: n_merges = 151387 print_info: BOS token = 151643 '<|endoftext|>' print_info: EOS token = 151645 '<|im_end|>' print_info: EOT token = 151645 '<|im_end|>' print_info: PAD token = 151643 '<|endoftext|>' print_info: LF token = 198 'Ċ' print_info: FIM PRE token = 151659 '<|fim_prefix|>' print_info: FIM SUF token = 151661 '<|fim_suffix|>' print_info: FIM MID token = 151660 '<|fim_middle|>' print_info: FIM PAD token = 151662 '<|fim_pad|>' print_info: FIM REP token = 151663 '<|repo_name|>' print_info: FIM SEP token = 151664 '<|file_sep|>' print_info: EOG token = 151643 '<|endoftext|>' print_info: EOG token = 151645 '<|im_end|>' print_info: EOG token = 151662 '<|fim_pad|>' print_info: EOG token = 151663 '<|repo_name|>' print_info: EOG token = 151664 '<|file_sep|>' print_info: max token length = 256 llama_model_load: vocab only - skipping tensors time=2025-07-15T09:46:11.303+08:00 level=INFO source=server.go:438 msg="starting llama server" cmd="C:\\Users\\tts2\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --model C:\\Users\\tts2\\.ollama\\models\\blobs\\sha256-3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6 --ctx-size 8192 --batch-size 512 --threads 8 --no-mmap --parallel 2 --port 59254" time=2025-07-15T09:46:11.317+08:00 level=INFO source=sched.go:483 msg="loaded runners" count=1 time=2025-07-15T09:46:11.317+08:00 level=INFO source=server.go:598 msg="waiting for llama runner to start responding" time=2025-07-15T09:46:11.318+08:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server error" time=2025-07-15T09:46:11.357+08:00 level=INFO source=runner.go:815 msg="starting go runner" ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no ggml_cuda_init: found 1 CUDA devices: Device 0: AMD Radeon 780M Graphics [ZLUDA], compute capability 8.8, VMM: no load_backend: loaded CUDA backend from C:\Users\tts2\AppData\Local\Programs\Ollama\lib\ollama\ggml-cuda.dll load_backend: loaded CPU backend from C:\Users\tts2\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-icelake.dll time=2025-07-15T09:46:11.483+08:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.AVX512=1 CPU.0.AVX512_VBMI=1 CPU.0.AVX512_VNNI=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 CUDA.0.ARCHS=500,600,610,700,750,800,860,870,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(clang) time=2025-07-15T09:46:11.484+08:00 level=INFO source=runner.go:874 msg="Server listening on 127.0.0.1:59254" time=2025-07-15T09:46:11.569+08:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server loading model" llama_model_load_from_file_impl: using device CUDA0 (AMD Radeon 780M Graphics [ZLUDA]) - 12391 MiB free llama_model_loader: loaded meta data with 27 key-value pairs and 311 tensors from C:\Users\tts2\.ollama\models\blobs\sha256-3d0b790534fe4b79525fc3692950408dca41171676ed7e21db57af5c65ef6ab6 (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 = qwen3 llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.name str = Qwen3 1.7B llama_model_loader: - kv 3: general.basename str = Qwen3 llama_model_loader: - kv 4: general.size_label str = 1.7B llama_model_loader: - kv 5: qwen3.block_count u32 = 28 llama_model_loader: - kv 6: qwen3.context_length u32 = 40960 llama_model_loader: - kv 7: qwen3.embedding_length u32 = 2048 llama_model_loader: - kv 8: qwen3.feed_forward_length u32 = 6144 llama_model_loader: - kv 9: qwen3.attention.head_count u32 = 16 llama_model_loader: - kv 10: qwen3.attention.head_count_kv u32 = 8 llama_model_loader: - kv 11: qwen3.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 12: qwen3.attention.layer_norm_rms_epsilon f32 = 0.000001 llama_model_loader: - kv 13: qwen3.attention.key_length u32 = 128 llama_model_loader: - kv 14: qwen3.attention.value_length u32 = 128 llama_model_loader: - kv 15: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 16: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 17: tokenizer.ggml.tokens arr[str,151936] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 18: tokenizer.ggml.token_type arr[i32,151936] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 19: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 151645 llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 22: tokenizer.ggml.bos_token_id u32 = 151643 llama_model_loader: - kv 23: tokenizer.ggml.add_bos_token bool = false llama_model_loader: - kv 24: tokenizer.chat_template str = {%- if tools %}\n {{- '<|im_start|>... llama_model_loader: - kv 25: general.quantization_version u32 = 2 llama_model_loader: - kv 26: general.file_type u32 = 15 llama_model_loader: - type f32: 113 tensors llama_model_loader: - type f16: 28 tensors llama_model_loader: - type q4_K: 155 tensors llama_model_loader: - type q6_K: 15 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 1.26 GiB (5.33 BPW) load: special tokens cache size = 26 load: token to piece cache size = 0.9311 MB print_info: arch = qwen3 print_info: vocab_only = 0 print_info: n_ctx_train = 40960 print_info: n_embd = 2048 print_info: n_layer = 28 print_info: n_head = 16 print_info: n_head_kv = 8 print_info: n_rot = 128 print_info: n_swa = 0 print_info: n_swa_pattern = 1 print_info: n_embd_head_k = 128 print_info: n_embd_head_v = 128 print_info: n_gqa = 2 print_info: n_embd_k_gqa = 1024 print_info: n_embd_v_gqa = 1024 print_info: f_norm_eps = 0.0e+00 print_info: f_norm_rms_eps = 1.0e-06 print_info: f_clamp_kqv = 0.0e+00 print_info: f_max_alibi_bias = 0.0e+00 print_info: f_logit_scale = 0.0e+00 print_info: f_attn_scale = 0.0e+00 print_info: n_ff = 6144 print_info: n_expert = 0 print_info: n_expert_used = 0 print_info: causal attn = 1 print_info: pooling type = 0 print_info: rope type = 2 print_info: rope scaling = linear print_info: freq_base_train = 1000000.0 print_info: freq_scale_train = 1 print_info: n_ctx_orig_yarn = 40960 print_info: rope_finetuned = unknown print_info: ssm_d_conv = 0 print_info: ssm_d_inner = 0 print_info: ssm_d_state = 0 print_info: ssm_dt_rank = 0 print_info: ssm_dt_b_c_rms = 0 print_info: model type = 1.7B print_info: model params = 2.03 B print_info: general.name = Qwen3 1.7B print_info: vocab type = BPE print_info: n_vocab = 151936 print_info: n_merges = 151387 print_info: BOS token = 151643 '<|endoftext|>' print_info: EOS token = 151645 '<|im_end|>' print_info: EOT token = 151645 '<|im_end|>' print_info: PAD token = 151643 '<|endoftext|>' print_info: LF token = 198 'Ċ' print_info: FIM PRE token = 151659 '<|fim_prefix|>' print_info: FIM SUF token = 151661 '<|fim_suffix|>' print_info: FIM MID token = 151660 '<|fim_middle|>' print_info: FIM PAD token = 151662 '<|fim_pad|>' print_info: FIM REP token = 151663 '<|repo_name|>' print_info: FIM SEP token = 151664 '<|file_sep|>' print_info: EOG token = 151643 '<|endoftext|>' print_info: EOG token = 151645 '<|im_end|>' print_info: EOG token = 151662 '<|fim_pad|>' print_info: EOG token = 151663 '<|repo_name|>' print_info: EOG token = 151664 '<|file_sep|>' print_info: max token length = 256 load_tensors: loading model tensors, this can take a while... (mmap = false) load_tensors: offloading 0 repeating layers to GPU load_tensors: offloaded 0/29 layers to GPU load_tensors: CPU model buffer size = 166.92 MiB load_tensors: CUDA_Host model buffer size = 1123.71 MiB llama_context: constructing llama_context llama_context: n_seq_max = 2 llama_context: n_ctx = 8192 llama_context: n_ctx_per_seq = 4096 llama_context: n_batch = 1024 llama_context: n_ubatch = 512 llama_context: causal_attn = 1 llama_context: flash_attn = 0 llama_context: freq_base = 1000000.0 llama_context: freq_scale = 1 llama_context: n_ctx_per_seq (4096) < n_ctx_train (40960) -- the full capacity of the model will not be utilized llama_context: CPU output buffer size = 1.17 MiB llama_kv_cache_unified: kv_size = 8192, type_k = 'f16', type_v = 'f16', n_layer = 28, can_shift = 1, padding = 32 llama_kv_cache_unified: CPU KV buffer size = 896.00 MiB llama_kv_cache_unified: KV self size = 896.00 MiB, K (f16): 448.00 MiB, V (f16): 448.00 MiB llama_context: CUDA0 compute buffer size = 570.93 MiB llama_context: CUDA_Host compute buffer size = 20.01 MiB llama_context: graph nodes = 1070 llama_context: graph splits = 368 (with bs=512), 1 (with bs=1) time=2025-07-15T09:46:12.570+08:00 level=INFO source=server.go:637 msg="llama runner started in 1.25 seconds" [GIN] 2025/07/15 - 09:46:12 | 200 | 1.5786592s | 127.0.0.1 | POST "/api/generate" C:\a\ollama\ollama\ml\backend\ggml\ggml\src\ggml-cuda\ggml-cuda.cu:2641: GGML_ASSERT(stat == cudaSuccess) failed time=2025-07-15T09:46:19.445+08:00 level=ERROR source=server.go:807 msg="post predict" error="Post \"http://127.0.0.1:59254/completion\": read tcp 127.0.0.1:59256->127.0.0.1:59254: wsarecv: An existing connection was forcibly closed by the remote host." [GIN] 2025/07/15 - 09:46:19 | 200 | 293.5268ms | 127.0.0.1 | POST "/api/chat" time=2025-07-15T09:46:19.563+08:00 level=ERROR source=server.go:464 msg="llama runner terminated" error="exit status 0xc0000409" ``` ### OS Windows ### GPU AMD ### CPU AMD ### Ollama version 0.9.6
GiteaMirror added the bug label 2026-04-29 05:09:35 -05:00
GiteaMirror commented 2026-04-29 05:09:36 -05:00
Author
Owner
Copy Link

@iAndyHD3 commented on GitHub (Jul 18, 2025):

same thing with AMD 7800xt with ROCm on windows

<!-- gh-comment-id:3090353093 --> @iAndyHD3 commented on GitHub (Jul 18, 2025): same thing with AMD 7800xt with ROCm on windows
GiteaMirror commented 2026-04-29 05:09:37 -05:00
Author
Owner
Copy Link

@yuriishizawa commented on GitHub (Aug 24, 2025):

Same thing here with AMD 6650XT with ROCm on linux

<!-- gh-comment-id:3217742918 --> @yuriishizawa commented on GitHub (Aug 24, 2025): Same thing here with AMD 6650XT with ROCm on linux
GiteaMirror commented 2026-04-29 05:09:37 -05:00
Author
Owner
Copy Link

@laniakeaccx commented on GitHub (Aug 27, 2025):

Same thing here with AMD 6800XT with ROCm on linux

<!-- gh-comment-id:3228860850 --> @laniakeaccx commented on GitHub (Aug 27, 2025): Same thing here with AMD 6800XT with ROCm on linux
GiteaMirror commented 2026-04-29 05:09:38 -05:00
Author
Owner
Copy Link

@Qikahome commented on GitHub (Aug 27, 2025):

Win11 23H2 + Same gpu with 128G DDR5 memory, same problem;
maybe we should issues at likelovewant/ollama-for-amd?

<!-- gh-comment-id:3229185320 --> @Qikahome commented on GitHub (Aug 27, 2025): Win11 23H2 + Same gpu with 128G DDR5 memory, same problem; maybe we should issues at likelovewant/ollama-for-amd?
GiteaMirror commented 2026-04-29 05:09:38 -05:00
Author
Owner
Copy Link

@ayylmaonade commented on GitHub (Aug 27, 2025):

Can confirm I'm having the same issue running ollama w/ ROCm. I'm currently re-pulling my models to see if that helps. RX 7900 XTX, Arch Linux, Kernel 6.16.3.

Issue is resolved as of ROCm 6.4.3-2

<!-- gh-comment-id:3230132976 --> @ayylmaonade commented on GitHub (Aug 27, 2025): Can confirm I'm having the same issue running ollama w/ ROCm. I'm currently re-pulling my models to see if that helps. RX 7900 XTX, Arch Linux, Kernel 6.16.3. **Issue is resolved as of ROCm 6.4.3-2**
GiteaMirror commented 2026-04-29 05:09:39 -05:00
Author
Owner
Copy Link

@sebastianillges commented on GitHub (Aug 28, 2025):

I ran into the same problem this morning after upgrading the rocm libs from 6.4.1 to 6.4.3. A downgrade fixed it for me.

<!-- gh-comment-id:3232026551 --> @sebastianillges commented on GitHub (Aug 28, 2025): I ran into the same problem this morning after upgrading the rocm libs from 6.4.1 to 6.4.3. A downgrade fixed it for me.
GiteaMirror commented 2026-04-29 05:09:40 -05:00
Author
Owner
Copy Link

@Qikahome commented on GitHub (Aug 28, 2025):

I "solved" the problem by updating GPU driver to latest (2025/7/24), but the output speeds by using 780M(GPU) and 8700G(CPU) are almost the same.

<!-- gh-comment-id:3232037197 --> @Qikahome commented on GitHub (Aug 28, 2025): I "solved" the problem by updating GPU driver to latest (2025/7/24), but the output speeds by using 780M(GPU) and 8700G(CPU) are almost the same.
GiteaMirror commented 2026-04-29 05:09:40 -05:00
Author
Owner
Copy Link

@jeancf commented on GitHub (Aug 29, 2025):

I ran into the same problem this morning after upgrading the rocm libs from 6.4.1 to 6.4.3. A downgrade fixed it for me.

Same for me. I confirm going back to 6.4.1 for all rocm packages fixes the issue (arch linux, amd 6800xt).

<!-- gh-comment-id:3236464160 --> @jeancf commented on GitHub (Aug 29, 2025): > I ran into the same problem this morning after upgrading the rocm libs from 6.4.1 to 6.4.3. A downgrade fixed it for me. Same for me. I confirm going back to 6.4.1 for all rocm packages fixes the issue (arch linux, amd 6800xt).
GiteaMirror commented 2026-04-29 05:09:41 -05:00
Author
Owner
Copy Link

@momouekai commented on GitHub (Sep 7, 2025):

Try use older version of ollama (to v0.11.6).
It worked even with latest rocm library.

<!-- gh-comment-id:3263391235 --> @momouekai commented on GitHub (Sep 7, 2025): Try use older version of ollama (to v0.11.6). It worked even with latest rocm library.
GiteaMirror commented 2026-04-29 05:09:43 -05:00
Author
Owner
Copy Link

@chymchym1905 commented on GitHub (Mar 22, 2026):

I'm crashing with 5060ti 16gb on windows. ollama version 0.18.2. nvidia driver version 591.44

<!-- gh-comment-id:4106777513 --> @chymchym1905 commented on GitHub (Mar 22, 2026): I'm crashing with 5060ti 16gb on windows. ollama version 0.18.2. nvidia driver version 591.44
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#54053
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?