github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-07-16 05:05:37 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #11657] Context length greater than 8k, unable to run on GPU #7706

New Issue
Open
opened 2026-04-12 19:48:45 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-12 19:48:45 -05:00
Owner
Copy Link

Originally created by @metaforget on GitHub (Aug 4, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/11657

What is the issue?

When the Context length of the ollama win desktop application is greater than 8k, running the qwen3-coder will only run in CPU memory. My graphics card is RTX5060ti 16G. Does anyone have a similar problem

Relevant log output


time=2025-08-04T23:43:36.256+08:00 level=INFO source=app_windows.go:272 msg="starting Ollama" app=C:\Users\lyw\AppData\Local\Programs\Ollama version=0.10.1 OS=Windows/10.0.26100
time=2025-08-04T23:43:36.268+08:00 level=INFO source=app.go:223 msg="initialized tools registry" tool_count=4
time=2025-08-04T23:43:36.277+08:00 level=INFO source=app.go:238 msg="starting ollama server"
time=2025-08-04T23:43:37.338+08:00 level=INFO source=app.go:268 msg="starting ui server" port=10007
time=2025-08-04T23:43:38.283+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/models http.pattern="GET /api/v1/models" http.status=200 http.d=1.5446ms request_id=1754322218281804200 version=0.10.1
time=2025-08-04T23:43:38.302+08:00 level=ERROR source=ui.go:1293 msg="failed to show model details" error="model 'gemma3:4b' not found" model=gemma3:4b
time=2025-08-04T23:43:38.303+08:00 level=ERROR source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/model/gemma3:4b/capabilities http.pattern="GET /api/v1/model/{model}/capabilities" http.status=500 http.d=3.5146ms request_id=1754322218299564800 version=0.10.1
time=2025-08-04T23:43:38.768+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/me http.pattern="GET /api/v1/me" http.status=200 http.d=509.393ms request_id=1754322218259306300 version=0.10.1
time=2025-08-04T23:43:39.307+08:00 level=ERROR source=ui.go:1293 msg="failed to show model details" error="model 'gemma3:4b' not found" model=gemma3:4b
time=2025-08-04T23:43:39.307+08:00 level=ERROR source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/model/gemma3:4b/capabilities http.pattern="GET /api/v1/model/{model}/capabilities" http.status=500 http.d=1.029ms request_id=1754322219306029900 version=0.10.1
time=2025-08-04T23:43:39.846+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/chats http.pattern="GET /api/v1/chats" http.status=200 http.d=0s request_id=1754322219846364800 version=0.10.1
time=2025-08-04T23:43:40.303+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/chat/019875a7-d628-7dd2-badf-f36a781caddf http.pattern="GET /api/v1/chat/{id}" http.status=200 http.d=1.0774ms request_id=1754322220302206300 version=0.10.1
time=2025-08-04T23:43:40.338+08:00 level=INFO source=updater.go:252 msg="beginning update checker" interval=1h0m0s
time=2025-08-04T23:43:41.316+08:00 level=ERROR source=ui.go:1293 msg="failed to show model details" error="model 'gemma3:4b' not found" model=gemma3:4b
time=2025-08-04T23:43:41.316+08:00 level=ERROR source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/model/gemma3:4b/capabilities http.pattern="GET /api/v1/model/{model}/capabilities" http.status=500 http.d=1.5506ms request_id=1754322221314746800 version=0.10.1
time=2025-08-04T23:43:45.319+08:00 level=ERROR source=ui.go:1293 msg="failed to show model details" error="model 'gemma3:4b' not found" model=gemma3:4b
time=2025-08-04T23:43:45.319+08:00 level=ERROR source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/model/gemma3:4b/capabilities http.pattern="GET /api/v1/model/{model}/capabilities" http.status=500 http.d=1.2797ms request_id=1754322225318520400 version=0.10.1
time=2025-08-04T23:43:45.897+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/model/qwen3-coder/capabilities http.pattern="GET /api/v1/model/{model}/capabilities" http.status=200 http.d=42.6911ms request_id=1754322225854660300 version=0.10.1
time=2025-08-04T23:43:48.262+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/chat/019875a7-d628-7dd2-badf-f36a781caddf http.pattern="GET /api/v1/chat/{id}" http.status=200 http.d=505.3µs request_id=1754322228261550800 version=0.10.1
time=2025-08-04T23:43:48.946+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/settings http.pattern="GET /api/v1/settings" http.status=200 http.d=0s request_id=1754322228946136000 version=0.10.1
time=2025-08-04T23:43:49.220+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/me http.pattern="GET /api/v1/me" http.status=200 http.d=273.7882ms request_id=1754322228946657500 version=0.10.1
time=2025-08-04T23:43:50.580+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/chats http.pattern="GET /api/v1/chats" http.status=200 http.d=0s request_id=1754322230580393000 version=0.10.1
time=2025-08-04T23:43:50.581+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/models http.pattern="GET /api/v1/models" http.status=200 http.d=1.5883ms request_id=1754322230580393000 version=0.10.1
time=2025-08-04T23:43:50.752+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/chat/019875a7-d628-7dd2-badf-f36a781caddf http.pattern="GET /api/v1/chat/{id}" http.status=200 http.d=736.3µs request_id=1754322230751567900 version=0.10.1
time=2025-08-04T23:43:54.855+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/chats http.pattern="GET /api/v1/chats" http.status=200 http.d=0s request_id=1754322234855320400 version=0.10.1
time=2025-08-04T23:43:54.862+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/chat/019875c0-ede5-712f-8aae-a3d2faa9d2e2 http.pattern="GET /api/v1/chat/{id}" http.status=200 http.d=518.4µs request_id=1754322234861525800 version=0.10.1
time=2025-08-04T23:43:54.864+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/models http.pattern="GET /api/v1/models" http.status=200 http.d=2.7658ms request_id=1754322234862044200 version=0.10.1
time=2025-08-04T23:44:05.416+08:00 level=ERROR source=ui.go:988 msg="chat stream error" error="llama runner process has terminated: GGML_ASSERT(ctx->mem_buffer != NULL) failed"
time=2025-08-04T23:44:05.417+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=POST http.path=/api/v1/chat/new http.pattern="POST /api/v1/chat/{id}" http.status=200 http.d=10.5641759s request_id=1754322234853077800 version=0.10.1




time=2025-08-04T23:43:37.355+08:00 level=INFO source=routes.go:1238 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:16384 OLLAMA_DEBUG:INFO OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 OLLAMA_INTEL_GPU:false OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:D:\\Ollama\\models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:1 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES:]"
time=2025-08-04T23:43:37.369+08:00 level=INFO source=images.go:476 msg="total blobs: 12"
time=2025-08-04T23:43:37.369+08:00 level=INFO source=images.go:483 msg="total unused blobs removed: 0"
time=2025-08-04T23:43:37.370+08:00 level=INFO source=routes.go:1291 msg="Listening on 127.0.0.1:11434 (version 0.10.1)"
time=2025-08-04T23:43:37.370+08:00 level=INFO source=gpu.go:217 msg="looking for compatible GPUs"
time=2025-08-04T23:43:37.370+08:00 level=INFO source=gpu_windows.go:167 msg=packages count=1
time=2025-08-04T23:43:37.370+08:00 level=INFO source=gpu_windows.go:214 msg="" package=0 cores=6 efficiency=0 threads=12
time=2025-08-04T23:43:37.490+08:00 level=INFO source=types.go:130 msg="inference compute" id=GPU-29b7b6f8-c916-d2af-3d78-ae460f9cbdd3 library=cuda variant=v12 compute=12.0 driver=12.9 name="NVIDIA GeForce RTX 5060 Ti" total="15.9 GiB" available="14.7 GiB"
[GIN] 2025/08/04 - 23:43:37 | 200 |            0s |       127.0.0.1 | GET      "/"
[GIN] 2025/08/04 - 23:43:38 | 200 |      1.0407ms |       127.0.0.1 | GET      "/api/tags"
[GIN] 2025/08/04 - 23:43:38 | 404 |      2.3238ms |       127.0.0.1 | POST     "/api/show"
[GIN] 2025/08/04 - 23:43:39 | 404 |       1.029ms |       127.0.0.1 | POST     "/api/show"
[GIN] 2025/08/04 - 23:43:41 | 404 |       1.046ms |       127.0.0.1 | POST     "/api/show"
[GIN] 2025/08/04 - 23:43:45 | 404 |      1.2797ms |       127.0.0.1 | POST     "/api/show"
[GIN] 2025/08/04 - 23:43:45 | 200 |     41.6601ms |       127.0.0.1 | POST     "/api/show"
[GIN] 2025/08/04 - 23:43:50 | 200 |      1.0776ms |       127.0.0.1 | GET      "/api/tags"
[GIN] 2025/08/04 - 23:43:54 | 200 |      1.7181ms |       127.0.0.1 | GET      "/api/tags"
[GIN] 2025/08/04 - 23:43:54 | 200 |     42.7375ms |       127.0.0.1 | POST     "/api/show"
[GIN] 2025/08/04 - 23:43:54 | 200 |     35.7893ms |       127.0.0.1 | POST     "/api/show"
time=2025-08-04T23:43:55.025+08:00 level=INFO source=server.go:135 msg="system memory" total="15.9 GiB" free="10.1 GiB" free_swap="19.6 GiB"
time=2025-08-04T23:43:55.026+08:00 level=INFO source=server.go:175 msg=offload library=cuda layers.requested=-1 layers.model=49 layers.offload=28 layers.split="" memory.available="[13.7 GiB]" memory.gpu_overhead="0 B" memory.required.full="21.5 GiB" memory.required.partial="13.6 GiB" memory.required.kv="1.5 GiB" memory.required.allocations="[13.6 GiB]" memory.weights.total="17.1 GiB" memory.weights.repeating="16.9 GiB" memory.weights.nonrepeating="243.4 MiB" memory.graph.full="2.0 GiB" memory.graph.partial="2.0 GiB"
llama_model_loader: loaded meta data with 35 key-value pairs and 579 tensors from D:\Ollama\models\blobs\sha256-1194192cf2a187eb02722edcc3f77b11d21f537048ce04b67ccf8ba78863006a (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              = qwen3moe
llama_model_loader: - kv   1:                           general.basename str              = Qwen3-Coder
llama_model_loader: - kv   2:                          general.file_type u32              = 15
llama_model_loader: - kv   3:                           general.finetune str              = Instruct
llama_model_loader: - kv   4:                            general.license str              = apache-2.0
llama_model_loader: - kv   5:                       general.license.link str              = https://huggingface.co/Qwen/Qwen3-Cod...
llama_model_loader: - kv   6:                               general.name str              = Qwen3 Coder 30B A3B Instruct
llama_model_loader: - kv   7:                    general.parameter_count u64              = 30532122624
llama_model_loader: - kv   8:               general.quantization_version u32              = 2
llama_model_loader: - kv   9:                         general.size_label str              = 30B-A3B
llama_model_loader: - kv  10:                               general.tags arr[str,1]       = ["text-generation"]
llama_model_loader: - kv  11:                               general.type str              = model
llama_model_loader: - kv  12:              qwen3moe.attention.head_count u32              = 32
llama_model_loader: - kv  13:           qwen3moe.attention.head_count_kv u32              = 4
llama_model_loader: - kv  14:              qwen3moe.attention.key_length u32              = 128
llama_model_loader: - kv  15:  qwen3moe.attention.layer_norm_rms_epsilon f32              = 0.000001
llama_model_loader: - kv  16:            qwen3moe.attention.value_length u32              = 128
llama_model_loader: - kv  17:                       qwen3moe.block_count u32              = 48
llama_model_loader: - kv  18:                    qwen3moe.context_length u32              = 262144
llama_model_loader: - kv  19:                  qwen3moe.embedding_length u32              = 2048
llama_model_loader: - kv  20:                      qwen3moe.expert_count u32              = 128
llama_model_loader: - kv  21:        qwen3moe.expert_feed_forward_length u32              = 768
llama_model_loader: - kv  22: qwen3moe.expert_shared_feed_forward_length u32              = 0
llama_model_loader: - kv  23:                 qwen3moe.expert_used_count u32              = 8
llama_model_loader: - kv  24:               qwen3moe.feed_forward_length u32              = 5472
llama_model_loader: - kv  25:                    qwen3moe.rope.freq_base f32              = 10000000.000000
llama_model_loader: - kv  26:                    tokenizer.chat_template str              = {% macro render_item_list(item_list, ...
llama_model_loader: - kv  27:               tokenizer.ggml.add_bos_token bool             = false
llama_model_loader: - kv  28:                tokenizer.ggml.eos_token_id u32              = 151645
llama_model_loader: - kv  29:                      tokenizer.ggml.merges arr[str,151387]  = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv  30:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  31:            tokenizer.ggml.padding_token_id u32              = 151643
llama_model_loader: - kv  32:                         tokenizer.ggml.pre str              = qwen2
llama_model_loader: - kv  33:                  tokenizer.ggml.token_type arr[i32,151936]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  34:                      tokenizer.ggml.tokens arr[str,151936]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - type  f32:  241 tensors
llama_model_loader: - type q4_K:  289 tensors
llama_model_loader: - type q6_K:   49 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q4_K - Medium
print_info: file size   = 17.28 GiB (4.86 BPW) 
load: special tokens cache size = 26
load: token to piece cache size = 0.9311 MB
print_info: arch             = qwen3moe
print_info: vocab_only       = 1
print_info: model type       = ?B
print_info: model params     = 30.53 B
print_info: general.name     = Qwen3 Coder 30B A3B Instruct
print_info: n_ff_exp         = 0
print_info: vocab type       = BPE
print_info: n_vocab          = 151936
print_info: n_merges         = 151387
print_info: BOS token        = 11 ','
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=2025-08-04T23:43:55.182+08:00 level=INFO source=server.go:438 msg="starting llama server" cmd="C:\\Users\\lyw\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --model D:\\Ollama\\models\\blobs\\sha256-1194192cf2a187eb02722edcc3f77b11d21f537048ce04b67ccf8ba78863006a --ctx-size 16384 --batch-size 512 --n-gpu-layers 28 --threads 6 --no-mmap --parallel 1 --port 10040"
time=2025-08-04T23:43:55.195+08:00 level=INFO source=sched.go:481 msg="loaded runners" count=1
time=2025-08-04T23:43:55.195+08:00 level=INFO source=server.go:598 msg="waiting for llama runner to start responding"
time=2025-08-04T23:43:55.195+08:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server error"
time=2025-08-04T23:43:55.225+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: NVIDIA GeForce RTX 5060 Ti, compute capability 12.0, VMM: yes
load_backend: loaded CUDA backend from C:\Users\lyw\AppData\Local\Programs\Ollama\lib\ollama\ggml-cuda.dll
load_backend: loaded CPU backend from C:\Users\lyw\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll
time=2025-08-04T23:43:55.330+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.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=2025-08-04T23:43:55.330+08:00 level=INFO source=runner.go:874 msg="Server listening on 127.0.0.1:10040"
llama_model_load_from_file_impl: using device CUDA0 (NVIDIA GeForce RTX 5060 Ti) - 15072 MiB free
llama_model_loader: loaded meta data with 35 key-value pairs and 579 tensors from D:\Ollama\models\blobs\sha256-1194192cf2a187eb02722edcc3f77b11d21f537048ce04b67ccf8ba78863006a (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              = qwen3moe
llama_model_loader: - kv   1:                           general.basename str              = Qwen3-Coder
llama_model_loader: - kv   2:                          general.file_type u32              = 15
llama_model_loader: - kv   3:                           general.finetune str              = Instruct
llama_model_loader: - kv   4:                            general.license str              = apache-2.0
llama_model_loader: - kv   5:                       general.license.link str              = https://huggingface.co/Qwen/Qwen3-Cod...
llama_model_loader: - kv   6:                               general.name str              = Qwen3 Coder 30B A3B Instruct
llama_model_loader: - kv   7:                    general.parameter_count u64              = 30532122624
llama_model_loader: - kv   8:               general.quantization_version u32              = 2
llama_model_loader: - kv   9:                         general.size_label str              = 30B-A3B
llama_model_loader: - kv  10:                               general.tags arr[str,1]       = ["text-generation"]
llama_model_loader: - kv  11:                               general.type str              = model
llama_model_loader: - kv  12:              qwen3moe.attention.head_count u32              = 32
llama_model_loader: - kv  13:           qwen3moe.attention.head_count_kv u32              = 4
llama_model_loader: - kv  14:              qwen3moe.attention.key_length u32              = 128
llama_model_loader: - kv  15:  qwen3moe.attention.layer_norm_rms_epsilon f32              = 0.000001
llama_model_loader: - kv  16:            qwen3moe.attention.value_length u32              = 128
llama_model_loader: - kv  17:                       qwen3moe.block_count u32              = 48
llama_model_loader: - kv  18:                    qwen3moe.context_length u32              = 262144
llama_model_loader: - kv  19:                  qwen3moe.embedding_length u32              = 2048
llama_model_loader: - kv  20:                      qwen3moe.expert_count u32              = 128
llama_model_loader: - kv  21:        qwen3moe.expert_feed_forward_length u32              = 768
llama_model_loader: - kv  22: qwen3moe.expert_shared_feed_forward_length u32              = 0
llama_model_loader: - kv  23:                 qwen3moe.expert_used_count u32              = 8
llama_model_loader: - kv  24:               qwen3moe.feed_forward_length u32              = 5472
llama_model_loader: - kv  25:                    qwen3moe.rope.freq_base f32              = 10000000.000000
llama_model_loader: - kv  26:                    tokenizer.chat_template str              = {% macro render_item_list(item_list, ...
llama_model_loader: - kv  27:               tokenizer.ggml.add_bos_token bool             = false
llama_model_loader: - kv  28:                tokenizer.ggml.eos_token_id u32              = 151645
llama_model_loader: - kv  29:                      tokenizer.ggml.merges arr[str,151387]  = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv  30:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  31:            tokenizer.ggml.padding_token_id u32              = 151643
llama_model_loader: - kv  32:                         tokenizer.ggml.pre str              = qwen2
llama_model_loader: - kv  33:                  tokenizer.ggml.token_type arr[i32,151936]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
time=2025-08-04T23:43:55.446+08:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server loading model"
llama_model_loader: - kv  34:                      tokenizer.ggml.tokens arr[str,151936]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - type  f32:  241 tensors
llama_model_loader: - type q4_K:  289 tensors
llama_model_loader: - type q6_K:   49 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q4_K - Medium
print_info: file size   = 17.28 GiB (4.86 BPW) 
load: special tokens cache size = 26
load: token to piece cache size = 0.9311 MB
print_info: arch             = qwen3moe
print_info: vocab_only       = 0
print_info: n_ctx_train      = 262144
print_info: n_embd           = 2048
print_info: n_layer          = 48
print_info: n_head           = 32
print_info: n_head_kv        = 4
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            = 8
print_info: n_embd_k_gqa     = 512
print_info: n_embd_v_gqa     = 512
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             = 5472
print_info: n_expert         = 128
print_info: n_expert_used    = 8
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  = 10000000.0
print_info: freq_scale_train = 1
print_info: n_ctx_orig_yarn  = 262144
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       = 30B.A3B
print_info: model params     = 30.53 B
print_info: general.name     = Qwen3 Coder 30B A3B Instruct
print_info: n_ff_exp         = 768
print_info: vocab type       = BPE
print_info: n_vocab          = 151936
print_info: n_merges         = 151387
print_info: BOS token        = 11 ','
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)
ggml_cuda_host_malloc: failed to allocate 7443.85 MiB of pinned memory: out of memory
load_tensors: offloading 28 repeating layers to GPU
load_tensors: offloaded 28/49 layers to GPU
load_tensors:          CPU model buffer size =  7443.85 MiB
load_tensors:        CUDA0 model buffer size = 10080.58 MiB
load_tensors:          CPU model buffer size =   166.92 MiB
llama_context: constructing llama_context
llama_context: n_seq_max     = 1
llama_context: n_ctx         = 16384
llama_context: n_ctx_per_seq = 16384
llama_context: n_batch       = 512
llama_context: n_ubatch      = 512
llama_context: causal_attn   = 1
llama_context: flash_attn    = 0
llama_context: freq_base     = 10000000.0
llama_context: freq_scale    = 1
llama_context: n_ctx_per_seq (16384) < n_ctx_train (262144) -- the full capacity of the model will not be utilized
llama_context:        CPU  output buffer size =     0.59 MiB
llama_kv_cache_unified: kv_size = 16384, type_k = 'f16', type_v = 'f16', n_layer = 48, can_shift = 1, padding = 32
llama_kv_cache_unified:      CUDA0 KV buffer size =   896.00 MiB
llama_kv_cache_unified:        CPU KV buffer size =   640.00 MiB
llama_kv_cache_unified: KV self size  = 1536.00 MiB, K (f16):  768.00 MiB, V (f16):  768.00 MiB
time=2025-08-04T23:44:04.660+08:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server not responding"
ggml.c:1422: GGML_ASSERT(ctx->mem_buffer != NULL) failed
time=2025-08-04T23:44:04.916+08:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server error"
time=2025-08-04T23:44:05.335+08:00 level=ERROR source=server.go:464 msg="llama runner terminated" error="exit status 0xc0000409"
time=2025-08-04T23:44:05.416+08:00 level=ERROR source=sched.go:487 msg="error loading llama server" error="llama runner process has terminated: GGML_ASSERT(ctx->mem_buffer != NULL) failed"
[GIN] 2025/08/04 - 23:44:05 | 500 |   10.4823286s |       127.0.0.1 | POST     "/api/chat"

OS

Windows

GPU

Nvidia

CPU

AMD

Ollama version

0.10.1

Originally created by @metaforget on GitHub (Aug 4, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/11657 ### What is the issue? When the Context length of the ollama win desktop application is greater than 8k, running the qwen3-coder will only run in CPU memory. My graphics card is RTX5060ti 16G. Does anyone have a similar problem ### Relevant log output ```shell time=2025-08-04T23:43:36.256+08:00 level=INFO source=app_windows.go:272 msg="starting Ollama" app=C:\Users\lyw\AppData\Local\Programs\Ollama version=0.10.1 OS=Windows/10.0.26100 time=2025-08-04T23:43:36.268+08:00 level=INFO source=app.go:223 msg="initialized tools registry" tool_count=4 time=2025-08-04T23:43:36.277+08:00 level=INFO source=app.go:238 msg="starting ollama server" time=2025-08-04T23:43:37.338+08:00 level=INFO source=app.go:268 msg="starting ui server" port=10007 time=2025-08-04T23:43:38.283+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/models http.pattern="GET /api/v1/models" http.status=200 http.d=1.5446ms request_id=1754322218281804200 version=0.10.1 time=2025-08-04T23:43:38.302+08:00 level=ERROR source=ui.go:1293 msg="failed to show model details" error="model 'gemma3:4b' not found" model=gemma3:4b time=2025-08-04T23:43:38.303+08:00 level=ERROR source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/model/gemma3:4b/capabilities http.pattern="GET /api/v1/model/{model}/capabilities" http.status=500 http.d=3.5146ms request_id=1754322218299564800 version=0.10.1 time=2025-08-04T23:43:38.768+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/me http.pattern="GET /api/v1/me" http.status=200 http.d=509.393ms request_id=1754322218259306300 version=0.10.1 time=2025-08-04T23:43:39.307+08:00 level=ERROR source=ui.go:1293 msg="failed to show model details" error="model 'gemma3:4b' not found" model=gemma3:4b time=2025-08-04T23:43:39.307+08:00 level=ERROR source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/model/gemma3:4b/capabilities http.pattern="GET /api/v1/model/{model}/capabilities" http.status=500 http.d=1.029ms request_id=1754322219306029900 version=0.10.1 time=2025-08-04T23:43:39.846+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/chats http.pattern="GET /api/v1/chats" http.status=200 http.d=0s request_id=1754322219846364800 version=0.10.1 time=2025-08-04T23:43:40.303+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/chat/019875a7-d628-7dd2-badf-f36a781caddf http.pattern="GET /api/v1/chat/{id}" http.status=200 http.d=1.0774ms request_id=1754322220302206300 version=0.10.1 time=2025-08-04T23:43:40.338+08:00 level=INFO source=updater.go:252 msg="beginning update checker" interval=1h0m0s time=2025-08-04T23:43:41.316+08:00 level=ERROR source=ui.go:1293 msg="failed to show model details" error="model 'gemma3:4b' not found" model=gemma3:4b time=2025-08-04T23:43:41.316+08:00 level=ERROR source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/model/gemma3:4b/capabilities http.pattern="GET /api/v1/model/{model}/capabilities" http.status=500 http.d=1.5506ms request_id=1754322221314746800 version=0.10.1 time=2025-08-04T23:43:45.319+08:00 level=ERROR source=ui.go:1293 msg="failed to show model details" error="model 'gemma3:4b' not found" model=gemma3:4b time=2025-08-04T23:43:45.319+08:00 level=ERROR source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/model/gemma3:4b/capabilities http.pattern="GET /api/v1/model/{model}/capabilities" http.status=500 http.d=1.2797ms request_id=1754322225318520400 version=0.10.1 time=2025-08-04T23:43:45.897+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/model/qwen3-coder/capabilities http.pattern="GET /api/v1/model/{model}/capabilities" http.status=200 http.d=42.6911ms request_id=1754322225854660300 version=0.10.1 time=2025-08-04T23:43:48.262+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/chat/019875a7-d628-7dd2-badf-f36a781caddf http.pattern="GET /api/v1/chat/{id}" http.status=200 http.d=505.3µs request_id=1754322228261550800 version=0.10.1 time=2025-08-04T23:43:48.946+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/settings http.pattern="GET /api/v1/settings" http.status=200 http.d=0s request_id=1754322228946136000 version=0.10.1 time=2025-08-04T23:43:49.220+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/me http.pattern="GET /api/v1/me" http.status=200 http.d=273.7882ms request_id=1754322228946657500 version=0.10.1 time=2025-08-04T23:43:50.580+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/chats http.pattern="GET /api/v1/chats" http.status=200 http.d=0s request_id=1754322230580393000 version=0.10.1 time=2025-08-04T23:43:50.581+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/models http.pattern="GET /api/v1/models" http.status=200 http.d=1.5883ms request_id=1754322230580393000 version=0.10.1 time=2025-08-04T23:43:50.752+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/chat/019875a7-d628-7dd2-badf-f36a781caddf http.pattern="GET /api/v1/chat/{id}" http.status=200 http.d=736.3µs request_id=1754322230751567900 version=0.10.1 time=2025-08-04T23:43:54.855+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/chats http.pattern="GET /api/v1/chats" http.status=200 http.d=0s request_id=1754322234855320400 version=0.10.1 time=2025-08-04T23:43:54.862+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/chat/019875c0-ede5-712f-8aae-a3d2faa9d2e2 http.pattern="GET /api/v1/chat/{id}" http.status=200 http.d=518.4µs request_id=1754322234861525800 version=0.10.1 time=2025-08-04T23:43:54.864+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=GET http.path=/api/v1/models http.pattern="GET /api/v1/models" http.status=200 http.d=2.7658ms request_id=1754322234862044200 version=0.10.1 time=2025-08-04T23:44:05.416+08:00 level=ERROR source=ui.go:988 msg="chat stream error" error="llama runner process has terminated: GGML_ASSERT(ctx->mem_buffer != NULL) failed" time=2025-08-04T23:44:05.417+08:00 level=INFO source=ui.go:162 msg=site.serveHTTP http.method=POST http.path=/api/v1/chat/new http.pattern="POST /api/v1/chat/{id}" http.status=200 http.d=10.5641759s request_id=1754322234853077800 version=0.10.1 time=2025-08-04T23:43:37.355+08:00 level=INFO source=routes.go:1238 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:16384 OLLAMA_DEBUG:INFO OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 OLLAMA_INTEL_GPU:false OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:D:\\Ollama\\models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:1 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES:]" time=2025-08-04T23:43:37.369+08:00 level=INFO source=images.go:476 msg="total blobs: 12" time=2025-08-04T23:43:37.369+08:00 level=INFO source=images.go:483 msg="total unused blobs removed: 0" time=2025-08-04T23:43:37.370+08:00 level=INFO source=routes.go:1291 msg="Listening on 127.0.0.1:11434 (version 0.10.1)" time=2025-08-04T23:43:37.370+08:00 level=INFO source=gpu.go:217 msg="looking for compatible GPUs" time=2025-08-04T23:43:37.370+08:00 level=INFO source=gpu_windows.go:167 msg=packages count=1 time=2025-08-04T23:43:37.370+08:00 level=INFO source=gpu_windows.go:214 msg="" package=0 cores=6 efficiency=0 threads=12 time=2025-08-04T23:43:37.490+08:00 level=INFO source=types.go:130 msg="inference compute" id=GPU-29b7b6f8-c916-d2af-3d78-ae460f9cbdd3 library=cuda variant=v12 compute=12.0 driver=12.9 name="NVIDIA GeForce RTX 5060 Ti" total="15.9 GiB" available="14.7 GiB" [GIN] 2025/08/04 - 23:43:37 | 200 | 0s | 127.0.0.1 | GET "/" [GIN] 2025/08/04 - 23:43:38 | 200 | 1.0407ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/08/04 - 23:43:38 | 404 | 2.3238ms | 127.0.0.1 | POST "/api/show" [GIN] 2025/08/04 - 23:43:39 | 404 | 1.029ms | 127.0.0.1 | POST "/api/show" [GIN] 2025/08/04 - 23:43:41 | 404 | 1.046ms | 127.0.0.1 | POST "/api/show" [GIN] 2025/08/04 - 23:43:45 | 404 | 1.2797ms | 127.0.0.1 | POST "/api/show" [GIN] 2025/08/04 - 23:43:45 | 200 | 41.6601ms | 127.0.0.1 | POST "/api/show" [GIN] 2025/08/04 - 23:43:50 | 200 | 1.0776ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/08/04 - 23:43:54 | 200 | 1.7181ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/08/04 - 23:43:54 | 200 | 42.7375ms | 127.0.0.1 | POST "/api/show" [GIN] 2025/08/04 - 23:43:54 | 200 | 35.7893ms | 127.0.0.1 | POST "/api/show" time=2025-08-04T23:43:55.025+08:00 level=INFO source=server.go:135 msg="system memory" total="15.9 GiB" free="10.1 GiB" free_swap="19.6 GiB" time=2025-08-04T23:43:55.026+08:00 level=INFO source=server.go:175 msg=offload library=cuda layers.requested=-1 layers.model=49 layers.offload=28 layers.split="" memory.available="[13.7 GiB]" memory.gpu_overhead="0 B" memory.required.full="21.5 GiB" memory.required.partial="13.6 GiB" memory.required.kv="1.5 GiB" memory.required.allocations="[13.6 GiB]" memory.weights.total="17.1 GiB" memory.weights.repeating="16.9 GiB" memory.weights.nonrepeating="243.4 MiB" memory.graph.full="2.0 GiB" memory.graph.partial="2.0 GiB" llama_model_loader: loaded meta data with 35 key-value pairs and 579 tensors from D:\Ollama\models\blobs\sha256-1194192cf2a187eb02722edcc3f77b11d21f537048ce04b67ccf8ba78863006a (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 = qwen3moe llama_model_loader: - kv 1: general.basename str = Qwen3-Coder llama_model_loader: - kv 2: general.file_type u32 = 15 llama_model_loader: - kv 3: general.finetune str = Instruct llama_model_loader: - kv 4: general.license str = apache-2.0 llama_model_loader: - kv 5: general.license.link str = https://huggingface.co/Qwen/Qwen3-Cod... llama_model_loader: - kv 6: general.name str = Qwen3 Coder 30B A3B Instruct llama_model_loader: - kv 7: general.parameter_count u64 = 30532122624 llama_model_loader: - kv 8: general.quantization_version u32 = 2 llama_model_loader: - kv 9: general.size_label str = 30B-A3B llama_model_loader: - kv 10: general.tags arr[str,1] = ["text-generation"] llama_model_loader: - kv 11: general.type str = model llama_model_loader: - kv 12: qwen3moe.attention.head_count u32 = 32 llama_model_loader: - kv 13: qwen3moe.attention.head_count_kv u32 = 4 llama_model_loader: - kv 14: qwen3moe.attention.key_length u32 = 128 llama_model_loader: - kv 15: qwen3moe.attention.layer_norm_rms_epsilon f32 = 0.000001 llama_model_loader: - kv 16: qwen3moe.attention.value_length u32 = 128 llama_model_loader: - kv 17: qwen3moe.block_count u32 = 48 llama_model_loader: - kv 18: qwen3moe.context_length u32 = 262144 llama_model_loader: - kv 19: qwen3moe.embedding_length u32 = 2048 llama_model_loader: - kv 20: qwen3moe.expert_count u32 = 128 llama_model_loader: - kv 21: qwen3moe.expert_feed_forward_length u32 = 768 llama_model_loader: - kv 22: qwen3moe.expert_shared_feed_forward_length u32 = 0 llama_model_loader: - kv 23: qwen3moe.expert_used_count u32 = 8 llama_model_loader: - kv 24: qwen3moe.feed_forward_length u32 = 5472 llama_model_loader: - kv 25: qwen3moe.rope.freq_base f32 = 10000000.000000 llama_model_loader: - kv 26: tokenizer.chat_template str = {% macro render_item_list(item_list, ... llama_model_loader: - kv 27: tokenizer.ggml.add_bos_token bool = false llama_model_loader: - kv 28: tokenizer.ggml.eos_token_id u32 = 151645 llama_model_loader: - kv 29: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 30: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 31: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 32: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 33: tokenizer.ggml.token_type arr[i32,151936] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 34: tokenizer.ggml.tokens arr[str,151936] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - type f32: 241 tensors llama_model_loader: - type q4_K: 289 tensors llama_model_loader: - type q6_K: 49 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 17.28 GiB (4.86 BPW) load: special tokens cache size = 26 load: token to piece cache size = 0.9311 MB print_info: arch = qwen3moe print_info: vocab_only = 1 print_info: model type = ?B print_info: model params = 30.53 B print_info: general.name = Qwen3 Coder 30B A3B Instruct print_info: n_ff_exp = 0 print_info: vocab type = BPE print_info: n_vocab = 151936 print_info: n_merges = 151387 print_info: BOS token = 11 ',' 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=2025-08-04T23:43:55.182+08:00 level=INFO source=server.go:438 msg="starting llama server" cmd="C:\\Users\\lyw\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --model D:\\Ollama\\models\\blobs\\sha256-1194192cf2a187eb02722edcc3f77b11d21f537048ce04b67ccf8ba78863006a --ctx-size 16384 --batch-size 512 --n-gpu-layers 28 --threads 6 --no-mmap --parallel 1 --port 10040" time=2025-08-04T23:43:55.195+08:00 level=INFO source=sched.go:481 msg="loaded runners" count=1 time=2025-08-04T23:43:55.195+08:00 level=INFO source=server.go:598 msg="waiting for llama runner to start responding" time=2025-08-04T23:43:55.195+08:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server error" time=2025-08-04T23:43:55.225+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: NVIDIA GeForce RTX 5060 Ti, compute capability 12.0, VMM: yes load_backend: loaded CUDA backend from C:\Users\lyw\AppData\Local\Programs\Ollama\lib\ollama\ggml-cuda.dll load_backend: loaded CPU backend from C:\Users\lyw\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll time=2025-08-04T23:43:55.330+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.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=2025-08-04T23:43:55.330+08:00 level=INFO source=runner.go:874 msg="Server listening on 127.0.0.1:10040" llama_model_load_from_file_impl: using device CUDA0 (NVIDIA GeForce RTX 5060 Ti) - 15072 MiB free llama_model_loader: loaded meta data with 35 key-value pairs and 579 tensors from D:\Ollama\models\blobs\sha256-1194192cf2a187eb02722edcc3f77b11d21f537048ce04b67ccf8ba78863006a (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 = qwen3moe llama_model_loader: - kv 1: general.basename str = Qwen3-Coder llama_model_loader: - kv 2: general.file_type u32 = 15 llama_model_loader: - kv 3: general.finetune str = Instruct llama_model_loader: - kv 4: general.license str = apache-2.0 llama_model_loader: - kv 5: general.license.link str = https://huggingface.co/Qwen/Qwen3-Cod... llama_model_loader: - kv 6: general.name str = Qwen3 Coder 30B A3B Instruct llama_model_loader: - kv 7: general.parameter_count u64 = 30532122624 llama_model_loader: - kv 8: general.quantization_version u32 = 2 llama_model_loader: - kv 9: general.size_label str = 30B-A3B llama_model_loader: - kv 10: general.tags arr[str,1] = ["text-generation"] llama_model_loader: - kv 11: general.type str = model llama_model_loader: - kv 12: qwen3moe.attention.head_count u32 = 32 llama_model_loader: - kv 13: qwen3moe.attention.head_count_kv u32 = 4 llama_model_loader: - kv 14: qwen3moe.attention.key_length u32 = 128 llama_model_loader: - kv 15: qwen3moe.attention.layer_norm_rms_epsilon f32 = 0.000001 llama_model_loader: - kv 16: qwen3moe.attention.value_length u32 = 128 llama_model_loader: - kv 17: qwen3moe.block_count u32 = 48 llama_model_loader: - kv 18: qwen3moe.context_length u32 = 262144 llama_model_loader: - kv 19: qwen3moe.embedding_length u32 = 2048 llama_model_loader: - kv 20: qwen3moe.expert_count u32 = 128 llama_model_loader: - kv 21: qwen3moe.expert_feed_forward_length u32 = 768 llama_model_loader: - kv 22: qwen3moe.expert_shared_feed_forward_length u32 = 0 llama_model_loader: - kv 23: qwen3moe.expert_used_count u32 = 8 llama_model_loader: - kv 24: qwen3moe.feed_forward_length u32 = 5472 llama_model_loader: - kv 25: qwen3moe.rope.freq_base f32 = 10000000.000000 llama_model_loader: - kv 26: tokenizer.chat_template str = {% macro render_item_list(item_list, ... llama_model_loader: - kv 27: tokenizer.ggml.add_bos_token bool = false llama_model_loader: - kv 28: tokenizer.ggml.eos_token_id u32 = 151645 llama_model_loader: - kv 29: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 30: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 31: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 32: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 33: tokenizer.ggml.token_type arr[i32,151936] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... time=2025-08-04T23:43:55.446+08:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server loading model" llama_model_loader: - kv 34: tokenizer.ggml.tokens arr[str,151936] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - type f32: 241 tensors llama_model_loader: - type q4_K: 289 tensors llama_model_loader: - type q6_K: 49 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 17.28 GiB (4.86 BPW) load: special tokens cache size = 26 load: token to piece cache size = 0.9311 MB print_info: arch = qwen3moe print_info: vocab_only = 0 print_info: n_ctx_train = 262144 print_info: n_embd = 2048 print_info: n_layer = 48 print_info: n_head = 32 print_info: n_head_kv = 4 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 = 8 print_info: n_embd_k_gqa = 512 print_info: n_embd_v_gqa = 512 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 = 5472 print_info: n_expert = 128 print_info: n_expert_used = 8 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 = 10000000.0 print_info: freq_scale_train = 1 print_info: n_ctx_orig_yarn = 262144 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 = 30B.A3B print_info: model params = 30.53 B print_info: general.name = Qwen3 Coder 30B A3B Instruct print_info: n_ff_exp = 768 print_info: vocab type = BPE print_info: n_vocab = 151936 print_info: n_merges = 151387 print_info: BOS token = 11 ',' 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) ggml_cuda_host_malloc: failed to allocate 7443.85 MiB of pinned memory: out of memory load_tensors: offloading 28 repeating layers to GPU load_tensors: offloaded 28/49 layers to GPU load_tensors: CPU model buffer size = 7443.85 MiB load_tensors: CUDA0 model buffer size = 10080.58 MiB load_tensors: CPU model buffer size = 166.92 MiB llama_context: constructing llama_context llama_context: n_seq_max = 1 llama_context: n_ctx = 16384 llama_context: n_ctx_per_seq = 16384 llama_context: n_batch = 512 llama_context: n_ubatch = 512 llama_context: causal_attn = 1 llama_context: flash_attn = 0 llama_context: freq_base = 10000000.0 llama_context: freq_scale = 1 llama_context: n_ctx_per_seq (16384) < n_ctx_train (262144) -- the full capacity of the model will not be utilized llama_context: CPU output buffer size = 0.59 MiB llama_kv_cache_unified: kv_size = 16384, type_k = 'f16', type_v = 'f16', n_layer = 48, can_shift = 1, padding = 32 llama_kv_cache_unified: CUDA0 KV buffer size = 896.00 MiB llama_kv_cache_unified: CPU KV buffer size = 640.00 MiB llama_kv_cache_unified: KV self size = 1536.00 MiB, K (f16): 768.00 MiB, V (f16): 768.00 MiB time=2025-08-04T23:44:04.660+08:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server not responding" ggml.c:1422: GGML_ASSERT(ctx->mem_buffer != NULL) failed time=2025-08-04T23:44:04.916+08:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server error" time=2025-08-04T23:44:05.335+08:00 level=ERROR source=server.go:464 msg="llama runner terminated" error="exit status 0xc0000409" time=2025-08-04T23:44:05.416+08:00 level=ERROR source=sched.go:487 msg="error loading llama server" error="llama runner process has terminated: GGML_ASSERT(ctx->mem_buffer != NULL) failed" [GIN] 2025/08/04 - 23:44:05 | 500 | 10.4823286s | 127.0.0.1 | POST "/api/chat" ``` ### OS Windows ### GPU Nvidia ### CPU AMD ### Ollama version 0.10.1
GiteaMirror added the bug label 2026-04-12 19:48:45 -05:00
GiteaMirror commented 2026-04-12 19:48:46 -05:00
Author
Owner
Copy Link

@metaforget commented on GitHub (Aug 4, 2025):

Image
<!-- gh-comment-id:3151304305 --> @metaforget commented on GitHub (Aug 4, 2025): <img width="1318" height="937" alt="Image" src="https://github.com/user-attachments/assets/693803ef-3fc8-4a99-9ecf-bdab3a2ac3bc" />
GiteaMirror commented 2026-04-12 19:48:46 -05:00
Author
Owner
Copy Link

@GoZippy commented on GitHub (Aug 4, 2025):

noticed behavior for last couple updates - models all run 100% cpu only presently and zero GPU. I have 4080.

<!-- gh-comment-id:3152521923 --> @GoZippy commented on GitHub (Aug 4, 2025): noticed behavior for last couple updates - models all run 100% cpu only presently and zero GPU. I have 4080.
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#7706
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?