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 #13083] GPU utilization issue #55172

New Issue
Closed
opened 2026-04-29 08:26:52 -05:00 by GiteaMirror · 5 comments
GiteaMirror commented 2026-04-29 08:26:52 -05:00
Owner
Copy Link

Originally created by @qihouji on GitHub (Nov 14, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/13083

After updating to version 0.12.11, GPU utilization dropped to 50%–60%.
System environment: Windows 11, CUDA 13 (RTX 5090 GPU).

Originally created by @qihouji on GitHub (Nov 14, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/13083 After updating to version 0.12.11, GPU utilization dropped to 50%–60%. System environment: Windows 11, CUDA 13 (RTX 5090 GPU).
GiteaMirror commented 2026-04-29 08:26:53 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 14, 2025):

Server log will help in debugging.

<!-- gh-comment-id:3531790285 --> @rick-github commented on GitHub (Nov 14, 2025): [Server log](https://docs.ollama.com/troubleshooting) will help in debugging.
GiteaMirror commented 2026-04-29 08:26:53 -05:00
Author
Owner
Copy Link

@moritzknecht commented on GitHub (Nov 15, 2025):

I also experience a performance regression on my linux system when upgrading to 0.12.11 from 0.12.10:

ollama version is 0.12.10
ollama run gpt-oss:20b-q4-128k --verbose hi
Thinking...
User says "hi". We need a friendly greeting. Probably ask how can I help.
...done thinking.

Hello! 👋 How can I help you today?

total duration: 2.842455436s
load duration: 2.628336154s
prompt eval count: 70 token(s)
prompt eval duration: 23.686579ms
prompt eval rate: 2955.26 tokens/s
eval count: 40 token(s)
eval duration: 161.480478ms
eval rate: 247.71 tokens/s

ollama version is 0.12.11
ollama run gpt-oss:20b-q4-128k --verbose hi
Thinking...
User says "hi". It's a greeting. Should respond politely. Probably ask how can help.
...done thinking.

Hello! 👋 How can I help you today?

total duration: 2.87789525s
load duration: 2.632629161s
prompt eval count: 70 token(s)
prompt eval duration: 23.924083ms
prompt eval rate: 2925.92 tokens/s
eval count: 40 token(s)
eval duration: 191.243912ms
eval rate: 209.16 tokens/s

----------------LOG 0.12.10-------------------

Nov 15 09:52:56 ai ollama[28485]: [GIN] 2025/11/15 - 09:52:56 | 200 | 18.907µs | 127.0.0.1 | HEAD "/"
Nov 15 09:52:56 ai ollama[28485]: time=2025-11-15T09:52:56.876Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.alignment default=32
Nov 15 09:52:56 ai ollama[28485]: [GIN] 2025/11/15 - 09:52:56 | 200 | 58.293681ms | 127.0.0.1 | POST "/api/show"
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.022Z level=DEBUG source=runner.go:243 msg="refreshing free memory"
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.022Z level=DEBUG source=runner.go:307 msg="unable to refresh all GPUs with existing runners, performing bootstrap discovery"
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.022Z level=INFO source=server.go:400 msg="starting runner" cmd="/usr/local/bin/ollama runner --ollama-engine --port 14257"
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.022Z level=DEBUG source=server.go:401 msg=subprocess PATH=/root/.opencode/bin:/usr/local/cuda-12.6/bin:/root/.nvm/versions/node/v22.18.0/bin:/root/.cargo/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin:/root/.local/bin:/usr/local/go/bin:/root/go/bin OLLAMA_HOST=0.0.0.0:11434 OLLAMA_NUM_THREADS=32 OLLAMA_NUM_PARALLEL=1 OLLAMA_MAX_LOADED_MODELS=1 OLLAMA_FLASH_ATTENTION=1 OLLAMA_KV_CACHE_TYPE=q8_0 OLLAMA_KEEP_ALIVE=60m OLLAMA_DEBUG=1 OLLAMA_USE_MLOCK=1 OLLAMA_NEW_ENGINE=1 OLLAMA_GPU_OVERHEAD=0 LD_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v12 OLLAMA_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v12
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.178Z level=DEBUG source=runner.go:415 msg="bootstrap discovery took" duration=156.625375ms OLLAMA_LIBRARY_PATH="[/usr/local/lib/ollama /usr/local/lib/ollama/cuda_v12]" extra_envs=map[]
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.178Z level=DEBUG source=runner.go:40 msg="overall device VRAM discovery took" duration=156.700899ms
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.197Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.alignment default=32
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.198Z level=DEBUG source=sched.go:211 msg="loading first model" model=/usr/share/ollama/.ollama/models/blobs/sha256-e7b273f9636059a689e3ddcab3716e4f65abe0143ac978e46673ad0e52d09efb
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.271Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.alignment default=32
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.271Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=gptoss.pooling_type default=0
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.271Z level=INFO source=server.go:215 msg="enabling flash attention"
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.271Z level=INFO source=server.go:400 msg="starting runner" cmd="/usr/local/bin/ollama runner --ollama-engine --model /usr/share/ollama/.ollama/models/blobs/sha256-e7b273f9636059a689e3ddcab3716e4f65abe0143ac978e46673ad0e52d09efb --port 37393"
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.271Z level=DEBUG source=server.go:401 msg=subprocess PATH=/root/.opencode/bin:/usr/local/cuda-12.6/bin:/root/.nvm/versions/node/v22.18.0/bin:/root/.cargo/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin:/root/.local/bin:/usr/local/go/bin:/root/go/bin OLLAMA_HOST=0.0.0.0:11434 OLLAMA_NUM_THREADS=32 OLLAMA_NUM_PARALLEL=1 OLLAMA_MAX_LOADED_MODELS=1 OLLAMA_FLASH_ATTENTION=1 OLLAMA_KV_CACHE_TYPE=q8_0 OLLAMA_KEEP_ALIVE=60m OLLAMA_DEBUG=1 OLLAMA_USE_MLOCK=1 OLLAMA_NEW_ENGINE=1 OLLAMA_GPU_OVERHEAD=0 LD_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v12 OLLAMA_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v12
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.271Z level=INFO source=server.go:653 msg="loading model" "model layers"=25 requested=-1
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.271Z level=INFO source=server.go:658 msg="system memory" total="46.8 GiB" free="43.0 GiB" free_swap="8.0 GiB"
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.271Z level=INFO source=server.go:665 msg="gpu memory" id=GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 library=CUDA available="30.9 GiB" free="31.3 GiB" minimum="457.0 MiB" overhead="0 B"
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.277Z level=INFO source=runner.go:1349 msg="starting ollama engine"
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.277Z level=INFO source=runner.go:1384 msg="Server listening on 127.0.0.1:37393"
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.283Z level=INFO source=runner.go:1222 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:1024 FlashAttention:true KvSize:131072 KvCacheType:q8_0 NumThreads:32 GPULayers:25[ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Layers:25(0..24)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.317Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.alignment default=32
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.317Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.name default=""
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.317Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.description default=""
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.317Z level=INFO source=ggml.go:136 msg="" architecture=gptoss file_type=MXFP4 name="" description="" num_tensors=459 num_key_values=32
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.317Z level=DEBUG source=ggml.go:94 msg="ggml backend load all from path" path=/usr/local/lib/ollama
Nov 15 09:52:57 ai ollama[28485]: load_backend: loaded CPU backend from /usr/local/lib/ollama/libggml-cpu-icelake.so
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.320Z level=DEBUG source=ggml.go:94 msg="ggml backend load all from path" path=/usr/local/lib/ollama/cuda_v12
Nov 15 09:52:57 ai ollama[28485]: ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no
Nov 15 09:52:57 ai ollama[28485]: ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
Nov 15 09:52:57 ai ollama[28485]: ggml_cuda_init: found 1 CUDA devices:
Nov 15 09:52:57 ai ollama[28485]: Device 0: NVIDIA GeForce RTX 5090, compute capability 12.0, VMM: yes, ID: GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9
Nov 15 09:52:57 ai ollama[28485]: load_backend: loaded CUDA backend from /usr/local/lib/ollama/cuda_v12/libggml-cuda.so
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.451Z 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=500,520,600,610,700,750,800,860,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(gcc)
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.452Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=gptoss.pooling_type default=0
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.906Z level=DEBUG source=ggml.go:853 msg="compute graph" nodes=1351 splits=2
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.915Z level=DEBUG source=ggml.go:853 msg="compute graph" nodes=1351 splits=2
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.915Z level=DEBUG source=device.go:212 msg="model weights" device=CUDA0 size="11.8 GiB"
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.915Z level=DEBUG source=device.go:217 msg="model weights" device=CPU size="1.1 GiB"
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.915Z level=DEBUG source=device.go:223 msg="kv cache" device=CUDA0 size="1.7 GiB"
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.915Z level=DEBUG source=device.go:234 msg="compute graph" device=CUDA0 size="1.1 GiB"
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.915Z level=DEBUG source=device.go:239 msg="compute graph" device=CPU size="11.2 MiB"
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.915Z level=DEBUG source=device.go:244 msg="total memory" size="15.6 GiB"
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.915Z level=DEBUG source=server.go:695 msg=memory success=true required.InputWeights=1158266880 required.CPU.Graph=11796480 required.CUDA0.ID=GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 required.CUDA0.Weights="[477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 1158278400]" required.CUDA0.Cache="[5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 0]" required.CUDA0.Graph=1198788736
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.915Z level=DEBUG source=server.go:892 msg="available gpu" id=GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 library=CUDA "available layer vram"="29.8 GiB" backoff=0.00 minimum="457.0 MiB" overhead="0 B" graph="1.1 GiB"
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.915Z level=DEBUG source=server.go:706 msg="new layout created" layers="25[ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Layers:25(0..24)]"
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.915Z level=INFO source=runner.go:1222 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:1024 FlashAttention:true KvSize:131072 KvCacheType:q8_0 NumThreads:32 GPULayers:25[ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Layers:25(0..24)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.944Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.alignment default=32
Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.946Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=gptoss.pooling_type default=0
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.473Z level=DEBUG source=ggml.go:853 msg="compute graph" nodes=1351 splits=2
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.502Z level=DEBUG source=ggml.go:853 msg="compute graph" nodes=1351 splits=2
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=DEBUG source=device.go:212 msg="model weights" device=CUDA0 size="11.8 GiB"
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=DEBUG source=device.go:217 msg="model weights" device=CPU size="1.1 GiB"
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=DEBUG source=device.go:223 msg="kv cache" device=CUDA0 size="1.7 GiB"
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=DEBUG source=device.go:234 msg="compute graph" device=CUDA0 size="1.1 GiB"
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=DEBUG source=device.go:239 msg="compute graph" device=CPU size="11.2 MiB"
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=DEBUG source=device.go:244 msg="total memory" size="15.6 GiB"
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=DEBUG source=server.go:695 msg=memory success=true required.InputWeights=1158266880 required.CPU.Graph=11796480 required.CUDA0.ID=GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 required.CUDA0.Weights="[477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 1158278400]" required.CUDA0.Cache="[5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 0]" required.CUDA0.Graph=1198788736
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=DEBUG source=server.go:892 msg="available gpu" id=GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 library=CUDA "available layer vram"="29.8 GiB" backoff=0.00 minimum="457.0 MiB" overhead="0 B" graph="1.1 GiB"
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=DEBUG source=server.go:706 msg="new layout created" layers="25[ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Layers:25(0..24)]"
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=INFO source=runner.go:1222 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:1024 FlashAttention:true KvSize:131072 KvCacheType:q8_0 NumThreads:32 GPULayers:25[ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Layers:25(0..24)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=INFO source=ggml.go:482 msg="offloading 24 repeating layers to GPU"
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=INFO source=ggml.go:489 msg="offloading output layer to GPU"
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=INFO source=ggml.go:494 msg="offloaded 25/25 layers to GPU"
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=INFO source=device.go:212 msg="model weights" device=CUDA0 size="11.8 GiB"
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=INFO source=device.go:217 msg="model weights" device=CPU size="1.1 GiB"
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=INFO source=device.go:223 msg="kv cache" device=CUDA0 size="1.7 GiB"
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=INFO source=device.go:234 msg="compute graph" device=CUDA0 size="1.1 GiB"
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=INFO source=device.go:239 msg="compute graph" device=CPU size="11.2 MiB"
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=INFO source=device.go:244 msg="total memory" size="15.6 GiB"
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=INFO source=sched.go:500 msg="loaded runners" count=1
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=INFO source=server.go:1251 msg="waiting for llama runner to start responding"
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.504Z level=INFO source=server.go:1285 msg="waiting for server to become available" status="llm server loading model"
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.504Z level=DEBUG source=server.go:1295 msg="model load progress 0.00"
Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.754Z level=DEBUG source=server.go:1295 msg="model load progress 0.27"
Nov 15 09:52:59 ai ollama[28485]: time=2025-11-15T09:52:59.005Z level=DEBUG source=server.go:1295 msg="model load progress 0.54"
Nov 15 09:52:59 ai ollama[28485]: time=2025-11-15T09:52:59.256Z level=DEBUG source=server.go:1295 msg="model load progress 0.83"
Nov 15 09:52:59 ai ollama[28485]: time=2025-11-15T09:52:59.496Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=gptoss.pooling_type default=0
Nov 15 09:52:59 ai ollama[28485]: time=2025-11-15T09:52:59.507Z level=INFO source=server.go:1289 msg="llama runner started in 2.24 seconds"
Nov 15 09:52:59 ai ollama[28485]: time=2025-11-15T09:52:59.507Z level=DEBUG source=sched.go:512 msg="finished setting up" runner.name=registry.ollama.ai/library/gpt-oss:20b-q4-128k runner.inference="[{ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Library:CUDA}]" runner.size="15.6 GiB" runner.vram="15.6 GiB" runner.parallel=1 runner.pid=28786 runner.model=/usr/share/ollama/.ollama/models/blobs/sha256-e7b273f9636059a689e3ddcab3716e4f65abe0143ac978e46673ad0e52d09efb runner.num_ctx=131072
Nov 15 09:52:59 ai ollama[28485]: time=2025-11-15T09:52:59.507Z level=DEBUG source=server.go:1401 msg="completion request" images=0 prompt=306 format=""
Nov 15 09:52:59 ai ollama[28485]: time=2025-11-15T09:52:59.527Z level=DEBUG source=cache.go:142 msg="loading cache slot" id=0 cache=0 prompt=70 used=0 remaining=70
Nov 15 09:52:59 ai ollama[28485]: time=2025-11-15T09:52:59.652Z level=WARN source=harmonyparser.go:122 msg="harmony parser: found message start tag in the middle of the content" content="\n<|start|>"
Nov 15 09:52:59 ai ollama[28485]: [GIN] 2025/11/15 - 09:52:59 | 200 | 2.870151156s | 127.0.0.1 | POST "/api/generate"
Nov 15 09:52:59 ai ollama[28485]: time=2025-11-15T09:52:59.748Z level=DEBUG source=sched.go:520 msg="context for request finished"
Nov 15 09:52:59 ai ollama[28485]: time=2025-11-15T09:52:59.748Z level=DEBUG source=sched.go:290 msg="runner with non-zero duration has gone idle, adding timer" runner.name=registry.ollama.ai/library/gpt-oss:20b-q4-128k runner.inference="[{ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Library:CUDA}]" runner.size="15.6 GiB" runner.vram="15.6 GiB" runner.parallel=1 runner.pid=28786 runner.model=/usr/share/ollama/.ollama/models/blobs/sha256-e7b273f9636059a689e3ddcab3716e4f65abe0143ac978e46673ad0e52d09efb runner.num_ctx=131072 duration=1h0m0s
Nov 15 09:52:59 ai ollama[28485]: time=2025-11-15T09:52:59.748Z level=DEBUG source=sched.go:308 msg="after processing request finished event" runner.name=registry.ollama.ai/library/gpt-oss:20b-q4-128k runner.inference="[{ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Library:CUDA}]" runner.size="15.6 GiB" runner.vram="15.6 GiB" runner.parallel=1 runner.pid=28786 runner.model=/usr/share/ollama/.ollama/models/blobs/sha256-e7b273f9636059a689e3ddcab3716e4f65abe0143ac978e46673ad0e52d09efb runner.num_ctx=131072 refCount=0

-----------------LOG 0.12.11----------------

Nov 15 09:51:22 ai ollama[27541]: [GIN] 2025/11/15 - 09:51:22 | 200 | 19.488µs | 127.0.0.1 | HEAD "/"
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.080Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.alignment default=32
Nov 15 09:51:22 ai ollama[27541]: [GIN] 2025/11/15 - 09:51:22 | 200 | 61.358602ms | 127.0.0.1 | POST "/api/show"
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.205Z level=DEBUG source=runner.go:246 msg="refreshing free memory"
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.205Z level=DEBUG source=runner.go:310 msg="unable to refresh all GPUs with existing runners, performing bootstrap discovery"
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.205Z level=INFO source=server.go:392 msg="starting runner" cmd="/usr/local/bin/ollama runner --ollama-engine --port 10781"
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.205Z level=DEBUG source=server.go:393 msg=subprocess PATH=/root/.opencode/bin:/usr/local/cuda-12.6/bin:/root/.nvm/versions/node/v22.18.0/bin:/root/.cargo/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin:/root/.local/bin:/usr/local/go/bin:/root/go/bin OLLAMA_HOST=0.0.0.0:11434 OLLAMA_NUM_THREADS=32 OLLAMA_NUM_PARALLEL=1 OLLAMA_MAX_LOADED_MODELS=1 OLLAMA_FLASH_ATTENTION=1 OLLAMA_KV_CACHE_TYPE=q8_0 OLLAMA_KEEP_ALIVE=60m OLLAMA_DEBUG=1 OLLAMA_USE_MLOCK=1 OLLAMA_NEW_ENGINE=1 OLLAMA_GPU_OVERHEAD=0 LD_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v12 OLLAMA_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v12
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.363Z level=DEBUG source=runner.go:418 msg="bootstrap discovery took" duration=158.218073ms OLLAMA_LIBRARY_PATH="[/usr/local/lib/ollama /usr/local/lib/ollama/cuda_v12]" extra_envs=map[]
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.363Z level=DEBUG source=runner.go:40 msg="overall device VRAM discovery took" duration=158.293037ms
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.381Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.alignment default=32
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.381Z level=DEBUG source=sched.go:211 msg="loading first model" model=/usr/share/ollama/.ollama/models/blobs/sha256-e7b273f9636059a689e3ddcab3716e4f65abe0143ac978e46673ad0e52d09efb
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.461Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.alignment default=32
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.461Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=gptoss.pooling_type default=0
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.461Z level=INFO source=server.go:209 msg="enabling flash attention"
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.461Z level=INFO source=server.go:392 msg="starting runner" cmd="/usr/local/bin/ollama runner --ollama-engine --model /usr/share/ollama/.ollama/models/blobs/sha256-e7b273f9636059a689e3ddcab3716e4f65abe0143ac978e46673ad0e52d09efb --port 11935"
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.461Z level=DEBUG source=server.go:393 msg=subprocess PATH=/root/.opencode/bin:/usr/local/cuda-12.6/bin:/root/.nvm/versions/node/v22.18.0/bin:/root/.cargo/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin:/root/.local/bin:/usr/local/go/bin:/root/go/bin OLLAMA_HOST=0.0.0.0:11434 OLLAMA_NUM_THREADS=32 OLLAMA_NUM_PARALLEL=1 OLLAMA_MAX_LOADED_MODELS=1 OLLAMA_FLASH_ATTENTION=1 OLLAMA_KV_CACHE_TYPE=q8_0 OLLAMA_KEEP_ALIVE=60m OLLAMA_DEBUG=1 OLLAMA_USE_MLOCK=1 OLLAMA_NEW_ENGINE=1 OLLAMA_GPU_OVERHEAD=0 LD_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v12 OLLAMA_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v12
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.462Z level=INFO source=sched.go:443 msg="system memory" total="46.8 GiB" free="43.0 GiB" free_swap="8.0 GiB"
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.462Z level=INFO source=sched.go:450 msg="gpu memory" id=GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 library=CUDA available="30.9 GiB" free="31.3 GiB" minimum="457.0 MiB" overhead="0 B"
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.462Z level=INFO source=server.go:702 msg="loading model" "model layers"=25 requested=-1
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.467Z level=INFO source=runner.go:1398 msg="starting ollama engine"
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.468Z level=INFO source=runner.go:1433 msg="Server listening on 127.0.0.1:11935"
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.473Z level=INFO source=runner.go:1271 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:1024 FlashAttention:true KvSize:131072 KvCacheType:q8_0 NumThreads:32 GPULayers:25[ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Layers:25(0..24)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.506Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.alignment default=32
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.507Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.name default=""
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.507Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.description default=""
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.507Z level=INFO source=ggml.go:136 msg="" architecture=gptoss file_type=MXFP4 name="" description="" num_tensors=459 num_key_values=32
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.507Z level=DEBUG source=ggml.go:94 msg="ggml backend load all from path" path=/usr/local/lib/ollama
Nov 15 09:51:22 ai ollama[27541]: load_backend: loaded CPU backend from /usr/local/lib/ollama/libggml-cpu-icelake.so
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.509Z level=DEBUG source=ggml.go:94 msg="ggml backend load all from path" path=/usr/local/lib/ollama/cuda_v12
Nov 15 09:51:22 ai ollama[27541]: ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no
Nov 15 09:51:22 ai ollama[27541]: ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
Nov 15 09:51:22 ai ollama[27541]: ggml_cuda_init: found 1 CUDA devices:
Nov 15 09:51:22 ai ollama[27541]: Device 0: NVIDIA GeForce RTX 5090, compute capability 12.0, VMM: yes, ID: GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9
Nov 15 09:51:22 ai ollama[27541]: load_backend: loaded CUDA backend from /usr/local/lib/ollama/cuda_v12/libggml-cuda.so
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.642Z 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=500,520,600,610,700,750,800,860,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(gcc)
Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.643Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=gptoss.pooling_type default=0
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.099Z level=DEBUG source=ggml.go:853 msg="compute graph" nodes=1351 splits=2
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.108Z level=DEBUG source=ggml.go:853 msg="compute graph" nodes=1351 splits=2
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.109Z level=DEBUG source=device.go:240 msg="model weights" device=CUDA0 size="11.8 GiB"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.109Z level=DEBUG source=device.go:245 msg="model weights" device=CPU size="1.1 GiB"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.109Z level=DEBUG source=device.go:251 msg="kv cache" device=CUDA0 size="1.7 GiB"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.109Z level=DEBUG source=device.go:262 msg="compute graph" device=CUDA0 size="1.1 GiB"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.109Z level=DEBUG source=device.go:267 msg="compute graph" device=CPU size="11.2 MiB"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.109Z level=DEBUG source=device.go:272 msg="total memory" size="15.6 GiB"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.109Z level=DEBUG source=server.go:727 msg=memory success=true required.InputWeights=1158266880 required.CPU.Graph=11796480 required.CUDA0.ID=GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 required.CUDA0.Weights="[477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 1158278400]" required.CUDA0.Cache="[5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 0]" required.CUDA0.Graph=1198788736
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.109Z level=DEBUG source=server.go:921 msg="available gpu" id=GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 library=CUDA "available layer vram"="29.8 GiB" backoff=0.00 minimum="457.0 MiB" overhead="0 B" graph="1.1 GiB"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.109Z level=DEBUG source=server.go:738 msg="new layout created" layers="25[ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Layers:25(0..24)]"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.109Z level=INFO source=runner.go:1271 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:1024 FlashAttention:true KvSize:131072 KvCacheType:q8_0 NumThreads:32 GPULayers:25[ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Layers:25(0..24)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.139Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.alignment default=32
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.140Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=gptoss.pooling_type default=0
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.683Z level=DEBUG source=ggml.go:853 msg="compute graph" nodes=1351 splits=2
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.711Z level=DEBUG source=ggml.go:853 msg="compute graph" nodes=1351 splits=2
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=DEBUG source=device.go:240 msg="model weights" device=CUDA0 size="11.8 GiB"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=DEBUG source=device.go:245 msg="model weights" device=CPU size="1.1 GiB"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=DEBUG source=device.go:251 msg="kv cache" device=CUDA0 size="1.7 GiB"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=DEBUG source=device.go:262 msg="compute graph" device=CUDA0 size="1.1 GiB"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=DEBUG source=device.go:267 msg="compute graph" device=CPU size="11.2 MiB"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=DEBUG source=device.go:272 msg="total memory" size="15.6 GiB"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=DEBUG source=server.go:727 msg=memory success=true required.InputWeights=1158266880 required.CPU.Graph=11796480 required.CUDA0.ID=GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 required.CUDA0.Weights="[477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 1158278400]" required.CUDA0.Cache="[5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 0]" required.CUDA0.Graph=1198788736
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=DEBUG source=server.go:921 msg="available gpu" id=GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 library=CUDA "available layer vram"="29.8 GiB" backoff=0.00 minimum="457.0 MiB" overhead="0 B" graph="1.1 GiB"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=DEBUG source=server.go:738 msg="new layout created" layers="25[ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Layers:25(0..24)]"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=runner.go:1271 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:1024 FlashAttention:true KvSize:131072 KvCacheType:q8_0 NumThreads:32 GPULayers:25[ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Layers:25(0..24)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=device.go:240 msg="model weights" device=CUDA0 size="11.8 GiB"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=device.go:245 msg="model weights" device=CPU size="1.1 GiB"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=device.go:251 msg="kv cache" device=CUDA0 size="1.7 GiB"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=device.go:262 msg="compute graph" device=CUDA0 size="1.1 GiB"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=device.go:267 msg="compute graph" device=CPU size="11.2 MiB"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=device.go:272 msg="total memory" size="15.6 GiB"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=ggml.go:482 msg="offloading 24 repeating layers to GPU"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=ggml.go:489 msg="offloading output layer to GPU"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=ggml.go:494 msg="offloaded 25/25 layers to GPU"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=sched.go:517 msg="loaded runners" count=1
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=server.go:1294 msg="waiting for llama runner to start responding"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=server.go:1328 msg="waiting for server to become available" status="llm server loading model"
Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.963Z level=DEBUG source=server.go:1338 msg="model load progress 0.27"
Nov 15 09:51:24 ai ollama[27541]: time=2025-11-15T09:51:24.214Z level=DEBUG source=server.go:1338 msg="model load progress 0.55"
Nov 15 09:51:24 ai ollama[27541]: time=2025-11-15T09:51:24.465Z level=DEBUG source=server.go:1338 msg="model load progress 0.83"
Nov 15 09:51:24 ai ollama[27541]: time=2025-11-15T09:51:24.697Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=gptoss.pooling_type default=0
Nov 15 09:51:24 ai ollama[27541]: time=2025-11-15T09:51:24.715Z level=INFO source=server.go:1332 msg="llama runner started in 2.25 seconds"
Nov 15 09:51:24 ai ollama[27541]: time=2025-11-15T09:51:24.715Z level=DEBUG source=sched.go:529 msg="finished setting up" runner.name=registry.ollama.ai/library/gpt-oss:20b-q4-128k runner.inference="[{ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Library:CUDA}]" runner.size="15.6 GiB" runner.vram="15.6 GiB" runner.parallel=1 runner.pid=27874 runner.model=/usr/share/ollama/.ollama/models/blobs/sha256-e7b273f9636059a689e3ddcab3716e4f65abe0143ac978e46673ad0e52d09efb runner.num_ctx=131072
Nov 15 09:51:24 ai ollama[27541]: time=2025-11-15T09:51:24.715Z level=DEBUG source=server.go:1465 msg="completion request" images=0 prompt=306 format=""
Nov 15 09:51:24 ai ollama[27541]: time=2025-11-15T09:51:24.736Z level=DEBUG source=cache.go:142 msg="loading cache slot" id=0 cache=0 prompt=70 used=0 remaining=70
Nov 15 09:51:24 ai ollama[27541]: [GIN] 2025/11/15 - 09:51:24 | 200 | 2.904959143s | 127.0.0.1 | POST "/api/generate"
Nov 15 09:51:24 ai ollama[27541]: time=2025-11-15T09:51:24.987Z level=DEBUG source=sched.go:537 msg="context for request finished"
Nov 15 09:51:24 ai ollama[27541]: time=2025-11-15T09:51:24.987Z level=DEBUG source=sched.go:290 msg="runner with non-zero duration has gone idle, adding timer" runner.name=registry.ollama.ai/library/gpt-oss:20b-q4-128k runner.inference="[{ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Library:CUDA}]" runner.size="15.6 GiB" runner.vram="15.6 GiB" runner.parallel=1 runner.pid=27874 runner.model=/usr/share/ollama/.ollama/models/blobs/sha256-e7b273f9636059a689e3ddcab3716e4f65abe0143ac978e46673ad0e52d09efb runner.num_ctx=131072 duration=1h0m0s
Nov 15 09:51:24 ai ollama[27541]: time=2025-11-15T09:51:24.987Z level=DEBUG source=sched.go:308 msg="after processing request finished event" runner.name=registry.ollama.ai/library/gpt-oss:20b-q4-128k runner.inference="[{ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Library:CUDA}]" runner.size="15.6 GiB" runner.vram="15.6 GiB" runner.parallel=1 runner.pid=27874 runner.model=/usr/share/ollama/.ollama/models/blobs/sha256-e7b273f9636059a689e3ddcab3716e4f65abe0143ac978e46673ad0e52d09efb runner.num_ctx=131072 refCount=0
Nov 15 09:52:35 ai systemd[1]: /etc/systemd/system/ollama.service.d/override.conf:9: Invalid syntax, ignoring: "OLLAMA_GPU_LAYERS=999""
Nov 15 09:52:35 ai systemd[1]: /etc/systemd/system/ollama.service.d/override.conf:9: Invalid syntax, ignoring: "OLLAMA_GPU_LAYERS=999""
Nov 15 09:52:35 ai systemd[1]: Stopping ollama.service - Ollama Service...
Nov 15 09:52:35 ai ollama[27541]: time=2025-11-15T09:52:35.753Z level=DEBUG source=sched.go:269 msg="shutting down scheduler completed loop"
Nov 15 09:52:35 ai ollama[27541]: time=2025-11-15T09:52:35.753Z level=DEBUG source=sched.go:844 msg="shutting down runner" model=/usr/share/ollama/.ollama/models/blobs/sha256-e7b273f9636059a689e3ddcab3716e4f65abe0143ac978e46673ad0e52d09efb
Nov 15 09:52:35 ai ollama[27541]: time=2025-11-15T09:52:35.753Z level=DEBUG source=sched.go:136 msg="shutting down scheduler pending loop"
Nov 15 09:52:35 ai ollama[27541]: time=2025-11-15T09:52:35.753Z level=DEBUG source=server.go:1755 msg="stopping llama server" pid=27874
Nov 15 09:52:35 ai ollama[27541]: time=2025-11-15T09:52:35.753Z level=DEBUG source=server.go:1761 msg="waiting for llama server to exit" pid=27874
Nov 15 09:52:36 ai ollama[27541]: time=2025-11-15T09:52:36.270Z level=DEBUG source=server.go:1765 msg="llama server stopped" pid=27874
Nov 15 09:52:36 ai systemd[1]: ollama.service: Deactivated successfully.
Nov 15 09:52:36 ai systemd[1]: Stopped ollama.service - Ollama Service.
Nov 15 09:52:36 ai systemd[1]: ollama.service: Consumed 6.048s CPU time, 2.7G memory peak, 0B memory swap peak.
Nov 15 09:52:36 ai systemd[1]: Started ollama.service - Ollama Service.
Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.305Z level=INFO source=routes.go:1525 msg="server config" env="map[CUDA_VISIBLE_DEVICES: GGML_VK_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:4096 OLLAMA_DEBUG:DEBUG OLLAMA_FLASH_ATTENTION:true OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://0.0.0.0:11434 OLLAMA_INTEL_GPU:false OLLAMA_KEEP_ALIVE:1h0m0s OLLAMA_KV_CACHE_TYPE:q8_0 OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:1 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/usr/share/ollama/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:true 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_REMOTES:[ollama.com] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES: http_proxy: https_proxy: no_proxy:]"
Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.311Z level=INFO source=images.go:522 msg="total blobs: 234"
Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.312Z level=INFO source=images.go:529 msg="total unused blobs removed: 0"
Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.313Z level=INFO source=routes.go:1578 msg="Listening on [::]:11434 (version 0.12.10)"
Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.313Z level=DEBUG source=sched.go:120 msg="starting llm scheduler"
Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.313Z level=INFO source=runner.go:67 msg="discovering available GPUs..."
Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.313Z level=INFO source=server.go:400 msg="starting runner" cmd="/usr/local/bin/ollama runner --ollama-engine --port 13649"
Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.313Z level=DEBUG source=server.go:401 msg=subprocess PATH=/root/.opencode/bin:/usr/local/cuda-12.6/bin:/root/.nvm/versions/node/v22.18.0/bin:/root/.cargo/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin:/root/.local/bin:/usr/local/go/bin:/root/go/bin OLLAMA_HOST=0.0.0.0:11434 OLLAMA_NUM_THREADS=32 OLLAMA_NUM_PARALLEL=1 OLLAMA_MAX_LOADED_MODELS=1 OLLAMA_FLASH_ATTENTION=1 OLLAMA_KV_CACHE_TYPE=q8_0 OLLAMA_KEEP_ALIVE=60m OLLAMA_DEBUG=1 OLLAMA_USE_MLOCK=1 OLLAMA_NEW_ENGINE=1 OLLAMA_GPU_OVERHEAD=0 LD_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v12 OLLAMA_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v12
Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.416Z level=DEBUG source=runner.go:415 msg="bootstrap discovery took" duration=102.96082ms OLLAMA_LIBRARY_PATH="[/usr/local/lib/ollama /usr/local/lib/ollama/cuda_v12]" extra_envs=map[]
Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.416Z level=INFO source=server.go:400 msg="starting runner" cmd="/usr/local/bin/ollama runner --ollama-engine --port 31709"
Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.416Z level=DEBUG source=server.go:401 msg=subprocess PATH=/root/.opencode/bin:/usr/local/cuda-12.6/bin:/root/.nvm/versions/node/v22.18.0/bin:/root/.cargo/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin:/root/.local/bin:/usr/local/go/bin:/root/go/bin OLLAMA_HOST=0.0.0.0:11434 OLLAMA_NUM_THREADS=32 OLLAMA_NUM_PARALLEL=1 OLLAMA_MAX_LOADED_MODELS=1 OLLAMA_FLASH_ATTENTION=1 OLLAMA_KV_CACHE_TYPE=q8_0 OLLAMA_KEEP_ALIVE=60m OLLAMA_DEBUG=1 OLLAMA_USE_MLOCK=1 OLLAMA_NEW_ENGINE=1 OLLAMA_GPU_OVERHEAD=0 LD_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v13 OLLAMA_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v13
Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.516Z level=DEBUG source=runner.go:415 msg="bootstrap discovery took" duration=99.668277ms OLLAMA_LIBRARY_PATH="[/usr/local/lib/ollama /usr/local/lib/ollama/cuda_v13]" extra_envs=map[]
Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.516Z level=DEBUG source=runner.go:113 msg="evluating which if any devices to filter out" initial_count=2
Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.516Z level=DEBUG source=runner.go:40 msg="GPU bootstrap discovery took" duration=202.830124ms
Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.516Z level=INFO source=types.go:42 msg="inference compute" id=GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 filter_id="" library=CUDA compute=12.0 name=CUDA0 description="NVIDIA GeForce RTX 5090" libdirs=ollama,cuda_v12 driver=13.0 pci_id=0000:01:00.0 type=discrete total="31.8 GiB" available="31.3 GiB"
Nov 15 09:52:36 ai ollama[28485]: [GIN] 2025/11/15 - 09:52:36 | 200 | 51.009µs | 127.0.0.1 | GET "/api/version"

<!-- gh-comment-id:3536287269 --> @moritzknecht commented on GitHub (Nov 15, 2025): I also experience a performance regression on my linux system when upgrading to 0.12.11 from 0.12.10: ollama version is 0.12.10 ollama run gpt-oss:20b-q4-128k --verbose hi Thinking... User says "hi". We need a friendly greeting. Probably ask how can I help. ...done thinking. Hello! 👋 How can I help you today? total duration: 2.842455436s load duration: 2.628336154s prompt eval count: 70 token(s) prompt eval duration: 23.686579ms prompt eval rate: 2955.26 tokens/s eval count: 40 token(s) eval duration: 161.480478ms eval rate: 247.71 tokens/s ollama version is 0.12.11 ollama run gpt-oss:20b-q4-128k --verbose hi Thinking... User says "hi". It's a greeting. Should respond politely. Probably ask how can help. ...done thinking. Hello! 👋 How can I help you today? total duration: 2.87789525s load duration: 2.632629161s prompt eval count: 70 token(s) prompt eval duration: 23.924083ms prompt eval rate: 2925.92 tokens/s eval count: 40 token(s) eval duration: 191.243912ms eval rate: 209.16 tokens/s ----------------LOG 0.12.10------------------- Nov 15 09:52:56 ai ollama[28485]: [GIN] 2025/11/15 - 09:52:56 | 200 | 18.907µs | 127.0.0.1 | HEAD "/" Nov 15 09:52:56 ai ollama[28485]: time=2025-11-15T09:52:56.876Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.alignment default=32 Nov 15 09:52:56 ai ollama[28485]: [GIN] 2025/11/15 - 09:52:56 | 200 | 58.293681ms | 127.0.0.1 | POST "/api/show" Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.022Z level=DEBUG source=runner.go:243 msg="refreshing free memory" Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.022Z level=DEBUG source=runner.go:307 msg="unable to refresh all GPUs with existing runners, performing bootstrap discovery" Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.022Z level=INFO source=server.go:400 msg="starting runner" cmd="/usr/local/bin/ollama runner --ollama-engine --port 14257" Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.022Z level=DEBUG source=server.go:401 msg=subprocess PATH=/root/.opencode/bin:/usr/local/cuda-12.6/bin:/root/.nvm/versions/node/v22.18.0/bin:/root/.cargo/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin:/root/.local/bin:/usr/local/go/bin:/root/go/bin OLLAMA_HOST=0.0.0.0:11434 OLLAMA_NUM_THREADS=32 OLLAMA_NUM_PARALLEL=1 OLLAMA_MAX_LOADED_MODELS=1 OLLAMA_FLASH_ATTENTION=1 OLLAMA_KV_CACHE_TYPE=q8_0 OLLAMA_KEEP_ALIVE=60m OLLAMA_DEBUG=1 OLLAMA_USE_MLOCK=1 OLLAMA_NEW_ENGINE=1 OLLAMA_GPU_OVERHEAD=0 LD_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v12 OLLAMA_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v12 Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.178Z level=DEBUG source=runner.go:415 msg="bootstrap discovery took" duration=156.625375ms OLLAMA_LIBRARY_PATH="[/usr/local/lib/ollama /usr/local/lib/ollama/cuda_v12]" extra_envs=map[] Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.178Z level=DEBUG source=runner.go:40 msg="overall device VRAM discovery took" duration=156.700899ms Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.197Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.alignment default=32 Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.198Z level=DEBUG source=sched.go:211 msg="loading first model" model=/usr/share/ollama/.ollama/models/blobs/sha256-e7b273f9636059a689e3ddcab3716e4f65abe0143ac978e46673ad0e52d09efb Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.271Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.alignment default=32 Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.271Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=gptoss.pooling_type default=0 Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.271Z level=INFO source=server.go:215 msg="enabling flash attention" Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.271Z level=INFO source=server.go:400 msg="starting runner" cmd="/usr/local/bin/ollama runner --ollama-engine --model /usr/share/ollama/.ollama/models/blobs/sha256-e7b273f9636059a689e3ddcab3716e4f65abe0143ac978e46673ad0e52d09efb --port 37393" Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.271Z level=DEBUG source=server.go:401 msg=subprocess PATH=/root/.opencode/bin:/usr/local/cuda-12.6/bin:/root/.nvm/versions/node/v22.18.0/bin:/root/.cargo/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin:/root/.local/bin:/usr/local/go/bin:/root/go/bin OLLAMA_HOST=0.0.0.0:11434 OLLAMA_NUM_THREADS=32 OLLAMA_NUM_PARALLEL=1 OLLAMA_MAX_LOADED_MODELS=1 OLLAMA_FLASH_ATTENTION=1 OLLAMA_KV_CACHE_TYPE=q8_0 OLLAMA_KEEP_ALIVE=60m OLLAMA_DEBUG=1 OLLAMA_USE_MLOCK=1 OLLAMA_NEW_ENGINE=1 OLLAMA_GPU_OVERHEAD=0 LD_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v12 OLLAMA_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v12 Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.271Z level=INFO source=server.go:653 msg="loading model" "model layers"=25 requested=-1 Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.271Z level=INFO source=server.go:658 msg="system memory" total="46.8 GiB" free="43.0 GiB" free_swap="8.0 GiB" Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.271Z level=INFO source=server.go:665 msg="gpu memory" id=GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 library=CUDA available="30.9 GiB" free="31.3 GiB" minimum="457.0 MiB" overhead="0 B" Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.277Z level=INFO source=runner.go:1349 msg="starting ollama engine" Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.277Z level=INFO source=runner.go:1384 msg="Server listening on 127.0.0.1:37393" Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.283Z level=INFO source=runner.go:1222 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:1024 FlashAttention:true KvSize:131072 KvCacheType:q8_0 NumThreads:32 GPULayers:25[ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Layers:25(0..24)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.317Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.alignment default=32 Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.317Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.name default="" Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.317Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.description default="" Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.317Z level=INFO source=ggml.go:136 msg="" architecture=gptoss file_type=MXFP4 name="" description="" num_tensors=459 num_key_values=32 Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.317Z level=DEBUG source=ggml.go:94 msg="ggml backend load all from path" path=/usr/local/lib/ollama Nov 15 09:52:57 ai ollama[28485]: load_backend: loaded CPU backend from /usr/local/lib/ollama/libggml-cpu-icelake.so Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.320Z level=DEBUG source=ggml.go:94 msg="ggml backend load all from path" path=/usr/local/lib/ollama/cuda_v12 Nov 15 09:52:57 ai ollama[28485]: ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no Nov 15 09:52:57 ai ollama[28485]: ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no Nov 15 09:52:57 ai ollama[28485]: ggml_cuda_init: found 1 CUDA devices: Nov 15 09:52:57 ai ollama[28485]: Device 0: NVIDIA GeForce RTX 5090, compute capability 12.0, VMM: yes, ID: GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Nov 15 09:52:57 ai ollama[28485]: load_backend: loaded CUDA backend from /usr/local/lib/ollama/cuda_v12/libggml-cuda.so Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.451Z 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=500,520,600,610,700,750,800,860,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(gcc) Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.452Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=gptoss.pooling_type default=0 Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.906Z level=DEBUG source=ggml.go:853 msg="compute graph" nodes=1351 splits=2 Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.915Z level=DEBUG source=ggml.go:853 msg="compute graph" nodes=1351 splits=2 Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.915Z level=DEBUG source=device.go:212 msg="model weights" device=CUDA0 size="11.8 GiB" Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.915Z level=DEBUG source=device.go:217 msg="model weights" device=CPU size="1.1 GiB" Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.915Z level=DEBUG source=device.go:223 msg="kv cache" device=CUDA0 size="1.7 GiB" Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.915Z level=DEBUG source=device.go:234 msg="compute graph" device=CUDA0 size="1.1 GiB" Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.915Z level=DEBUG source=device.go:239 msg="compute graph" device=CPU size="11.2 MiB" Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.915Z level=DEBUG source=device.go:244 msg="total memory" size="15.6 GiB" Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.915Z level=DEBUG source=server.go:695 msg=memory success=true required.InputWeights=1158266880 required.CPU.Graph=11796480 required.CUDA0.ID=GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 required.CUDA0.Weights="[477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 1158278400]" required.CUDA0.Cache="[5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 0]" required.CUDA0.Graph=1198788736 Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.915Z level=DEBUG source=server.go:892 msg="available gpu" id=GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 library=CUDA "available layer vram"="29.8 GiB" backoff=0.00 minimum="457.0 MiB" overhead="0 B" graph="1.1 GiB" Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.915Z level=DEBUG source=server.go:706 msg="new layout created" layers="25[ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Layers:25(0..24)]" Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.915Z level=INFO source=runner.go:1222 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:1024 FlashAttention:true KvSize:131072 KvCacheType:q8_0 NumThreads:32 GPULayers:25[ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Layers:25(0..24)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.944Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.alignment default=32 Nov 15 09:52:57 ai ollama[28485]: time=2025-11-15T09:52:57.946Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=gptoss.pooling_type default=0 Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.473Z level=DEBUG source=ggml.go:853 msg="compute graph" nodes=1351 splits=2 Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.502Z level=DEBUG source=ggml.go:853 msg="compute graph" nodes=1351 splits=2 Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=DEBUG source=device.go:212 msg="model weights" device=CUDA0 size="11.8 GiB" Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=DEBUG source=device.go:217 msg="model weights" device=CPU size="1.1 GiB" Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=DEBUG source=device.go:223 msg="kv cache" device=CUDA0 size="1.7 GiB" Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=DEBUG source=device.go:234 msg="compute graph" device=CUDA0 size="1.1 GiB" Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=DEBUG source=device.go:239 msg="compute graph" device=CPU size="11.2 MiB" Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=DEBUG source=device.go:244 msg="total memory" size="15.6 GiB" Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=DEBUG source=server.go:695 msg=memory success=true required.InputWeights=1158266880 required.CPU.Graph=11796480 required.CUDA0.ID=GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 required.CUDA0.Weights="[477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 1158278400]" required.CUDA0.Cache="[5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 0]" required.CUDA0.Graph=1198788736 Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=DEBUG source=server.go:892 msg="available gpu" id=GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 library=CUDA "available layer vram"="29.8 GiB" backoff=0.00 minimum="457.0 MiB" overhead="0 B" graph="1.1 GiB" Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=DEBUG source=server.go:706 msg="new layout created" layers="25[ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Layers:25(0..24)]" Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=INFO source=runner.go:1222 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:1024 FlashAttention:true KvSize:131072 KvCacheType:q8_0 NumThreads:32 GPULayers:25[ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Layers:25(0..24)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=INFO source=ggml.go:482 msg="offloading 24 repeating layers to GPU" Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=INFO source=ggml.go:489 msg="offloading output layer to GPU" Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=INFO source=ggml.go:494 msg="offloaded 25/25 layers to GPU" Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=INFO source=device.go:212 msg="model weights" device=CUDA0 size="11.8 GiB" Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=INFO source=device.go:217 msg="model weights" device=CPU size="1.1 GiB" Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=INFO source=device.go:223 msg="kv cache" device=CUDA0 size="1.7 GiB" Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=INFO source=device.go:234 msg="compute graph" device=CUDA0 size="1.1 GiB" Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=INFO source=device.go:239 msg="compute graph" device=CPU size="11.2 MiB" Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=INFO source=device.go:244 msg="total memory" size="15.6 GiB" Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=INFO source=sched.go:500 msg="loaded runners" count=1 Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.503Z level=INFO source=server.go:1251 msg="waiting for llama runner to start responding" Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.504Z level=INFO source=server.go:1285 msg="waiting for server to become available" status="llm server loading model" Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.504Z level=DEBUG source=server.go:1295 msg="model load progress 0.00" Nov 15 09:52:58 ai ollama[28485]: time=2025-11-15T09:52:58.754Z level=DEBUG source=server.go:1295 msg="model load progress 0.27" Nov 15 09:52:59 ai ollama[28485]: time=2025-11-15T09:52:59.005Z level=DEBUG source=server.go:1295 msg="model load progress 0.54" Nov 15 09:52:59 ai ollama[28485]: time=2025-11-15T09:52:59.256Z level=DEBUG source=server.go:1295 msg="model load progress 0.83" Nov 15 09:52:59 ai ollama[28485]: time=2025-11-15T09:52:59.496Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=gptoss.pooling_type default=0 Nov 15 09:52:59 ai ollama[28485]: time=2025-11-15T09:52:59.507Z level=INFO source=server.go:1289 msg="llama runner started in 2.24 seconds" Nov 15 09:52:59 ai ollama[28485]: time=2025-11-15T09:52:59.507Z level=DEBUG source=sched.go:512 msg="finished setting up" runner.name=registry.ollama.ai/library/gpt-oss:20b-q4-128k runner.inference="[{ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Library:CUDA}]" runner.size="15.6 GiB" runner.vram="15.6 GiB" runner.parallel=1 runner.pid=28786 runner.model=/usr/share/ollama/.ollama/models/blobs/sha256-e7b273f9636059a689e3ddcab3716e4f65abe0143ac978e46673ad0e52d09efb runner.num_ctx=131072 Nov 15 09:52:59 ai ollama[28485]: time=2025-11-15T09:52:59.507Z level=DEBUG source=server.go:1401 msg="completion request" images=0 prompt=306 format="" Nov 15 09:52:59 ai ollama[28485]: time=2025-11-15T09:52:59.527Z level=DEBUG source=cache.go:142 msg="loading cache slot" id=0 cache=0 prompt=70 used=0 remaining=70 Nov 15 09:52:59 ai ollama[28485]: time=2025-11-15T09:52:59.652Z level=WARN source=harmonyparser.go:122 msg="harmony parser: found message start tag in the middle of the content" content="\n<|start|>" Nov 15 09:52:59 ai ollama[28485]: [GIN] 2025/11/15 - 09:52:59 | 200 | 2.870151156s | 127.0.0.1 | POST "/api/generate" Nov 15 09:52:59 ai ollama[28485]: time=2025-11-15T09:52:59.748Z level=DEBUG source=sched.go:520 msg="context for request finished" Nov 15 09:52:59 ai ollama[28485]: time=2025-11-15T09:52:59.748Z level=DEBUG source=sched.go:290 msg="runner with non-zero duration has gone idle, adding timer" runner.name=registry.ollama.ai/library/gpt-oss:20b-q4-128k runner.inference="[{ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Library:CUDA}]" runner.size="15.6 GiB" runner.vram="15.6 GiB" runner.parallel=1 runner.pid=28786 runner.model=/usr/share/ollama/.ollama/models/blobs/sha256-e7b273f9636059a689e3ddcab3716e4f65abe0143ac978e46673ad0e52d09efb runner.num_ctx=131072 duration=1h0m0s Nov 15 09:52:59 ai ollama[28485]: time=2025-11-15T09:52:59.748Z level=DEBUG source=sched.go:308 msg="after processing request finished event" runner.name=registry.ollama.ai/library/gpt-oss:20b-q4-128k runner.inference="[{ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Library:CUDA}]" runner.size="15.6 GiB" runner.vram="15.6 GiB" runner.parallel=1 runner.pid=28786 runner.model=/usr/share/ollama/.ollama/models/blobs/sha256-e7b273f9636059a689e3ddcab3716e4f65abe0143ac978e46673ad0e52d09efb runner.num_ctx=131072 refCount=0 -----------------LOG 0.12.11---------------- Nov 15 09:51:22 ai ollama[27541]: [GIN] 2025/11/15 - 09:51:22 | 200 | 19.488µs | 127.0.0.1 | HEAD "/" Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.080Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.alignment default=32 Nov 15 09:51:22 ai ollama[27541]: [GIN] 2025/11/15 - 09:51:22 | 200 | 61.358602ms | 127.0.0.1 | POST "/api/show" Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.205Z level=DEBUG source=runner.go:246 msg="refreshing free memory" Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.205Z level=DEBUG source=runner.go:310 msg="unable to refresh all GPUs with existing runners, performing bootstrap discovery" Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.205Z level=INFO source=server.go:392 msg="starting runner" cmd="/usr/local/bin/ollama runner --ollama-engine --port 10781" Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.205Z level=DEBUG source=server.go:393 msg=subprocess PATH=/root/.opencode/bin:/usr/local/cuda-12.6/bin:/root/.nvm/versions/node/v22.18.0/bin:/root/.cargo/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin:/root/.local/bin:/usr/local/go/bin:/root/go/bin OLLAMA_HOST=0.0.0.0:11434 OLLAMA_NUM_THREADS=32 OLLAMA_NUM_PARALLEL=1 OLLAMA_MAX_LOADED_MODELS=1 OLLAMA_FLASH_ATTENTION=1 OLLAMA_KV_CACHE_TYPE=q8_0 OLLAMA_KEEP_ALIVE=60m OLLAMA_DEBUG=1 OLLAMA_USE_MLOCK=1 OLLAMA_NEW_ENGINE=1 OLLAMA_GPU_OVERHEAD=0 LD_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v12 OLLAMA_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v12 Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.363Z level=DEBUG source=runner.go:418 msg="bootstrap discovery took" duration=158.218073ms OLLAMA_LIBRARY_PATH="[/usr/local/lib/ollama /usr/local/lib/ollama/cuda_v12]" extra_envs=map[] Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.363Z level=DEBUG source=runner.go:40 msg="overall device VRAM discovery took" duration=158.293037ms Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.381Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.alignment default=32 Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.381Z level=DEBUG source=sched.go:211 msg="loading first model" model=/usr/share/ollama/.ollama/models/blobs/sha256-e7b273f9636059a689e3ddcab3716e4f65abe0143ac978e46673ad0e52d09efb Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.461Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.alignment default=32 Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.461Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=gptoss.pooling_type default=0 Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.461Z level=INFO source=server.go:209 msg="enabling flash attention" Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.461Z level=INFO source=server.go:392 msg="starting runner" cmd="/usr/local/bin/ollama runner --ollama-engine --model /usr/share/ollama/.ollama/models/blobs/sha256-e7b273f9636059a689e3ddcab3716e4f65abe0143ac978e46673ad0e52d09efb --port 11935" Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.461Z level=DEBUG source=server.go:393 msg=subprocess PATH=/root/.opencode/bin:/usr/local/cuda-12.6/bin:/root/.nvm/versions/node/v22.18.0/bin:/root/.cargo/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin:/root/.local/bin:/usr/local/go/bin:/root/go/bin OLLAMA_HOST=0.0.0.0:11434 OLLAMA_NUM_THREADS=32 OLLAMA_NUM_PARALLEL=1 OLLAMA_MAX_LOADED_MODELS=1 OLLAMA_FLASH_ATTENTION=1 OLLAMA_KV_CACHE_TYPE=q8_0 OLLAMA_KEEP_ALIVE=60m OLLAMA_DEBUG=1 OLLAMA_USE_MLOCK=1 OLLAMA_NEW_ENGINE=1 OLLAMA_GPU_OVERHEAD=0 LD_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v12 OLLAMA_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v12 Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.462Z level=INFO source=sched.go:443 msg="system memory" total="46.8 GiB" free="43.0 GiB" free_swap="8.0 GiB" Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.462Z level=INFO source=sched.go:450 msg="gpu memory" id=GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 library=CUDA available="30.9 GiB" free="31.3 GiB" minimum="457.0 MiB" overhead="0 B" Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.462Z level=INFO source=server.go:702 msg="loading model" "model layers"=25 requested=-1 Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.467Z level=INFO source=runner.go:1398 msg="starting ollama engine" Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.468Z level=INFO source=runner.go:1433 msg="Server listening on 127.0.0.1:11935" Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.473Z level=INFO source=runner.go:1271 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:1024 FlashAttention:true KvSize:131072 KvCacheType:q8_0 NumThreads:32 GPULayers:25[ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Layers:25(0..24)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.506Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.alignment default=32 Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.507Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.name default="" Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.507Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.description default="" Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.507Z level=INFO source=ggml.go:136 msg="" architecture=gptoss file_type=MXFP4 name="" description="" num_tensors=459 num_key_values=32 Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.507Z level=DEBUG source=ggml.go:94 msg="ggml backend load all from path" path=/usr/local/lib/ollama Nov 15 09:51:22 ai ollama[27541]: load_backend: loaded CPU backend from /usr/local/lib/ollama/libggml-cpu-icelake.so Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.509Z level=DEBUG source=ggml.go:94 msg="ggml backend load all from path" path=/usr/local/lib/ollama/cuda_v12 Nov 15 09:51:22 ai ollama[27541]: ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no Nov 15 09:51:22 ai ollama[27541]: ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no Nov 15 09:51:22 ai ollama[27541]: ggml_cuda_init: found 1 CUDA devices: Nov 15 09:51:22 ai ollama[27541]: Device 0: NVIDIA GeForce RTX 5090, compute capability 12.0, VMM: yes, ID: GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Nov 15 09:51:22 ai ollama[27541]: load_backend: loaded CUDA backend from /usr/local/lib/ollama/cuda_v12/libggml-cuda.so Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.642Z 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=500,520,600,610,700,750,800,860,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(gcc) Nov 15 09:51:22 ai ollama[27541]: time=2025-11-15T09:51:22.643Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=gptoss.pooling_type default=0 Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.099Z level=DEBUG source=ggml.go:853 msg="compute graph" nodes=1351 splits=2 Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.108Z level=DEBUG source=ggml.go:853 msg="compute graph" nodes=1351 splits=2 Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.109Z level=DEBUG source=device.go:240 msg="model weights" device=CUDA0 size="11.8 GiB" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.109Z level=DEBUG source=device.go:245 msg="model weights" device=CPU size="1.1 GiB" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.109Z level=DEBUG source=device.go:251 msg="kv cache" device=CUDA0 size="1.7 GiB" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.109Z level=DEBUG source=device.go:262 msg="compute graph" device=CUDA0 size="1.1 GiB" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.109Z level=DEBUG source=device.go:267 msg="compute graph" device=CPU size="11.2 MiB" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.109Z level=DEBUG source=device.go:272 msg="total memory" size="15.6 GiB" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.109Z level=DEBUG source=server.go:727 msg=memory success=true required.InputWeights=1158266880 required.CPU.Graph=11796480 required.CUDA0.ID=GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 required.CUDA0.Weights="[477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 1158278400]" required.CUDA0.Cache="[5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 0]" required.CUDA0.Graph=1198788736 Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.109Z level=DEBUG source=server.go:921 msg="available gpu" id=GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 library=CUDA "available layer vram"="29.8 GiB" backoff=0.00 minimum="457.0 MiB" overhead="0 B" graph="1.1 GiB" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.109Z level=DEBUG source=server.go:738 msg="new layout created" layers="25[ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Layers:25(0..24)]" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.109Z level=INFO source=runner.go:1271 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:1024 FlashAttention:true KvSize:131072 KvCacheType:q8_0 NumThreads:32 GPULayers:25[ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Layers:25(0..24)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.139Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=general.alignment default=32 Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.140Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=gptoss.pooling_type default=0 Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.683Z level=DEBUG source=ggml.go:853 msg="compute graph" nodes=1351 splits=2 Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.711Z level=DEBUG source=ggml.go:853 msg="compute graph" nodes=1351 splits=2 Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=DEBUG source=device.go:240 msg="model weights" device=CUDA0 size="11.8 GiB" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=DEBUG source=device.go:245 msg="model weights" device=CPU size="1.1 GiB" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=DEBUG source=device.go:251 msg="kv cache" device=CUDA0 size="1.7 GiB" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=DEBUG source=device.go:262 msg="compute graph" device=CUDA0 size="1.1 GiB" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=DEBUG source=device.go:267 msg="compute graph" device=CPU size="11.2 MiB" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=DEBUG source=device.go:272 msg="total memory" size="15.6 GiB" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=DEBUG source=server.go:727 msg=memory success=true required.InputWeights=1158266880 required.CPU.Graph=11796480 required.CUDA0.ID=GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 required.CUDA0.Weights="[477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 477628928 1158278400]" required.CUDA0.Cache="[5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 5571584 142607360 0]" required.CUDA0.Graph=1198788736 Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=DEBUG source=server.go:921 msg="available gpu" id=GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 library=CUDA "available layer vram"="29.8 GiB" backoff=0.00 minimum="457.0 MiB" overhead="0 B" graph="1.1 GiB" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=DEBUG source=server.go:738 msg="new layout created" layers="25[ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Layers:25(0..24)]" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=runner.go:1271 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:1024 FlashAttention:true KvSize:131072 KvCacheType:q8_0 NumThreads:32 GPULayers:25[ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Layers:25(0..24)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=device.go:240 msg="model weights" device=CUDA0 size="11.8 GiB" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=device.go:245 msg="model weights" device=CPU size="1.1 GiB" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=device.go:251 msg="kv cache" device=CUDA0 size="1.7 GiB" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=device.go:262 msg="compute graph" device=CUDA0 size="1.1 GiB" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=device.go:267 msg="compute graph" device=CPU size="11.2 MiB" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=device.go:272 msg="total memory" size="15.6 GiB" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=ggml.go:482 msg="offloading 24 repeating layers to GPU" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=ggml.go:489 msg="offloading output layer to GPU" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=ggml.go:494 msg="offloaded 25/25 layers to GPU" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=sched.go:517 msg="loaded runners" count=1 Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=server.go:1294 msg="waiting for llama runner to start responding" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.712Z level=INFO source=server.go:1328 msg="waiting for server to become available" status="llm server loading model" Nov 15 09:51:23 ai ollama[27541]: time=2025-11-15T09:51:23.963Z level=DEBUG source=server.go:1338 msg="model load progress 0.27" Nov 15 09:51:24 ai ollama[27541]: time=2025-11-15T09:51:24.214Z level=DEBUG source=server.go:1338 msg="model load progress 0.55" Nov 15 09:51:24 ai ollama[27541]: time=2025-11-15T09:51:24.465Z level=DEBUG source=server.go:1338 msg="model load progress 0.83" Nov 15 09:51:24 ai ollama[27541]: time=2025-11-15T09:51:24.697Z level=DEBUG source=ggml.go:276 msg="key with type not found" key=gptoss.pooling_type default=0 Nov 15 09:51:24 ai ollama[27541]: time=2025-11-15T09:51:24.715Z level=INFO source=server.go:1332 msg="llama runner started in 2.25 seconds" Nov 15 09:51:24 ai ollama[27541]: time=2025-11-15T09:51:24.715Z level=DEBUG source=sched.go:529 msg="finished setting up" runner.name=registry.ollama.ai/library/gpt-oss:20b-q4-128k runner.inference="[{ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Library:CUDA}]" runner.size="15.6 GiB" runner.vram="15.6 GiB" runner.parallel=1 runner.pid=27874 runner.model=/usr/share/ollama/.ollama/models/blobs/sha256-e7b273f9636059a689e3ddcab3716e4f65abe0143ac978e46673ad0e52d09efb runner.num_ctx=131072 Nov 15 09:51:24 ai ollama[27541]: time=2025-11-15T09:51:24.715Z level=DEBUG source=server.go:1465 msg="completion request" images=0 prompt=306 format="" Nov 15 09:51:24 ai ollama[27541]: time=2025-11-15T09:51:24.736Z level=DEBUG source=cache.go:142 msg="loading cache slot" id=0 cache=0 prompt=70 used=0 remaining=70 Nov 15 09:51:24 ai ollama[27541]: [GIN] 2025/11/15 - 09:51:24 | 200 | 2.904959143s | 127.0.0.1 | POST "/api/generate" Nov 15 09:51:24 ai ollama[27541]: time=2025-11-15T09:51:24.987Z level=DEBUG source=sched.go:537 msg="context for request finished" Nov 15 09:51:24 ai ollama[27541]: time=2025-11-15T09:51:24.987Z level=DEBUG source=sched.go:290 msg="runner with non-zero duration has gone idle, adding timer" runner.name=registry.ollama.ai/library/gpt-oss:20b-q4-128k runner.inference="[{ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Library:CUDA}]" runner.size="15.6 GiB" runner.vram="15.6 GiB" runner.parallel=1 runner.pid=27874 runner.model=/usr/share/ollama/.ollama/models/blobs/sha256-e7b273f9636059a689e3ddcab3716e4f65abe0143ac978e46673ad0e52d09efb runner.num_ctx=131072 duration=1h0m0s Nov 15 09:51:24 ai ollama[27541]: time=2025-11-15T09:51:24.987Z level=DEBUG source=sched.go:308 msg="after processing request finished event" runner.name=registry.ollama.ai/library/gpt-oss:20b-q4-128k runner.inference="[{ID:GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 Library:CUDA}]" runner.size="15.6 GiB" runner.vram="15.6 GiB" runner.parallel=1 runner.pid=27874 runner.model=/usr/share/ollama/.ollama/models/blobs/sha256-e7b273f9636059a689e3ddcab3716e4f65abe0143ac978e46673ad0e52d09efb runner.num_ctx=131072 refCount=0 Nov 15 09:52:35 ai systemd[1]: /etc/systemd/system/ollama.service.d/override.conf:9: Invalid syntax, ignoring: "OLLAMA_GPU_LAYERS=999"" Nov 15 09:52:35 ai systemd[1]: /etc/systemd/system/ollama.service.d/override.conf:9: Invalid syntax, ignoring: "OLLAMA_GPU_LAYERS=999"" Nov 15 09:52:35 ai systemd[1]: Stopping ollama.service - Ollama Service... Nov 15 09:52:35 ai ollama[27541]: time=2025-11-15T09:52:35.753Z level=DEBUG source=sched.go:269 msg="shutting down scheduler completed loop" Nov 15 09:52:35 ai ollama[27541]: time=2025-11-15T09:52:35.753Z level=DEBUG source=sched.go:844 msg="shutting down runner" model=/usr/share/ollama/.ollama/models/blobs/sha256-e7b273f9636059a689e3ddcab3716e4f65abe0143ac978e46673ad0e52d09efb Nov 15 09:52:35 ai ollama[27541]: time=2025-11-15T09:52:35.753Z level=DEBUG source=sched.go:136 msg="shutting down scheduler pending loop" Nov 15 09:52:35 ai ollama[27541]: time=2025-11-15T09:52:35.753Z level=DEBUG source=server.go:1755 msg="stopping llama server" pid=27874 Nov 15 09:52:35 ai ollama[27541]: time=2025-11-15T09:52:35.753Z level=DEBUG source=server.go:1761 msg="waiting for llama server to exit" pid=27874 Nov 15 09:52:36 ai ollama[27541]: time=2025-11-15T09:52:36.270Z level=DEBUG source=server.go:1765 msg="llama server stopped" pid=27874 Nov 15 09:52:36 ai systemd[1]: ollama.service: Deactivated successfully. Nov 15 09:52:36 ai systemd[1]: Stopped ollama.service - Ollama Service. Nov 15 09:52:36 ai systemd[1]: ollama.service: Consumed 6.048s CPU time, 2.7G memory peak, 0B memory swap peak. Nov 15 09:52:36 ai systemd[1]: Started ollama.service - Ollama Service. Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.305Z level=INFO source=routes.go:1525 msg="server config" env="map[CUDA_VISIBLE_DEVICES: GGML_VK_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:4096 OLLAMA_DEBUG:DEBUG OLLAMA_FLASH_ATTENTION:true OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://0.0.0.0:11434 OLLAMA_INTEL_GPU:false OLLAMA_KEEP_ALIVE:1h0m0s OLLAMA_KV_CACHE_TYPE:q8_0 OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:1 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/usr/share/ollama/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:true 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_REMOTES:[ollama.com] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES: http_proxy: https_proxy: no_proxy:]" Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.311Z level=INFO source=images.go:522 msg="total blobs: 234" Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.312Z level=INFO source=images.go:529 msg="total unused blobs removed: 0" Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.313Z level=INFO source=routes.go:1578 msg="Listening on [::]:11434 (version 0.12.10)" Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.313Z level=DEBUG source=sched.go:120 msg="starting llm scheduler" Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.313Z level=INFO source=runner.go:67 msg="discovering available GPUs..." Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.313Z level=INFO source=server.go:400 msg="starting runner" cmd="/usr/local/bin/ollama runner --ollama-engine --port 13649" Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.313Z level=DEBUG source=server.go:401 msg=subprocess PATH=/root/.opencode/bin:/usr/local/cuda-12.6/bin:/root/.nvm/versions/node/v22.18.0/bin:/root/.cargo/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin:/root/.local/bin:/usr/local/go/bin:/root/go/bin OLLAMA_HOST=0.0.0.0:11434 OLLAMA_NUM_THREADS=32 OLLAMA_NUM_PARALLEL=1 OLLAMA_MAX_LOADED_MODELS=1 OLLAMA_FLASH_ATTENTION=1 OLLAMA_KV_CACHE_TYPE=q8_0 OLLAMA_KEEP_ALIVE=60m OLLAMA_DEBUG=1 OLLAMA_USE_MLOCK=1 OLLAMA_NEW_ENGINE=1 OLLAMA_GPU_OVERHEAD=0 LD_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v12 OLLAMA_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v12 Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.416Z level=DEBUG source=runner.go:415 msg="bootstrap discovery took" duration=102.96082ms OLLAMA_LIBRARY_PATH="[/usr/local/lib/ollama /usr/local/lib/ollama/cuda_v12]" extra_envs=map[] Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.416Z level=INFO source=server.go:400 msg="starting runner" cmd="/usr/local/bin/ollama runner --ollama-engine --port 31709" Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.416Z level=DEBUG source=server.go:401 msg=subprocess PATH=/root/.opencode/bin:/usr/local/cuda-12.6/bin:/root/.nvm/versions/node/v22.18.0/bin:/root/.cargo/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin:/root/.local/bin:/usr/local/go/bin:/root/go/bin OLLAMA_HOST=0.0.0.0:11434 OLLAMA_NUM_THREADS=32 OLLAMA_NUM_PARALLEL=1 OLLAMA_MAX_LOADED_MODELS=1 OLLAMA_FLASH_ATTENTION=1 OLLAMA_KV_CACHE_TYPE=q8_0 OLLAMA_KEEP_ALIVE=60m OLLAMA_DEBUG=1 OLLAMA_USE_MLOCK=1 OLLAMA_NEW_ENGINE=1 OLLAMA_GPU_OVERHEAD=0 LD_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v13 OLLAMA_LIBRARY_PATH=/usr/local/lib/ollama:/usr/local/lib/ollama/cuda_v13 Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.516Z level=DEBUG source=runner.go:415 msg="bootstrap discovery took" duration=99.668277ms OLLAMA_LIBRARY_PATH="[/usr/local/lib/ollama /usr/local/lib/ollama/cuda_v13]" extra_envs=map[] Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.516Z level=DEBUG source=runner.go:113 msg="evluating which if any devices to filter out" initial_count=2 Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.516Z level=DEBUG source=runner.go:40 msg="GPU bootstrap discovery took" duration=202.830124ms Nov 15 09:52:36 ai ollama[28485]: time=2025-11-15T09:52:36.516Z level=INFO source=types.go:42 msg="inference compute" id=GPU-2794afa5-4f22-8cf6-43b9-279a053d98e9 filter_id="" library=CUDA compute=12.0 name=CUDA0 description="NVIDIA GeForce RTX 5090" libdirs=ollama,cuda_v12 driver=13.0 pci_id=0000:01:00.0 type=discrete total="31.8 GiB" available="31.3 GiB" Nov 15 09:52:36 ai ollama[28485]: [GIN] 2025/11/15 - 09:52:36 | 200 | 51.009µs | 127.0.0.1 | GET "/api/version"
GiteaMirror commented 2026-04-29 08:26:54 -05:00
Author
Owner
Copy Link

@qihouji commented on GitHub (Nov 19, 2025):

Startup log:
time=2025-11-19T20:26:35.460+08:00 level=INFO source=routes.go:1544 msg="server config" env="map[CUDA_VISIBLE_DEVICES:0 GGML_VK_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:4096 OLLAMA_DEBUG:INFO OLLAMA_FLASH_ATTENTION:true OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434d OLLAMA_KEEP_ALIVE:2562047h47m16.854775807s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:2 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:Q:\ollama_models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:2 OLLAMA_ORIGINS:[* http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_REMOTES:[ollama.com] OLLAMA_SCHED_SPREAD:false OLLAMA_VULKAN:false ROCR_VISIBLE_DEVICES:]"
time=2025-11-19T20:26:35.477+08:00 level=INFO source=images.go:522 msg="total blobs: 21"
time=2025-11-19T20:26:35.478+08:00 level=INFO source=images.go:529 msg="total unused blobs removed: 0"
time=2025-11-19T20:26:35.479+08:00 level=INFO source=routes.go:1597 msg="Listening on 127.0.0.1:11434 (version 0.12.11)"
time=2025-11-19T20:26:35.481+08:00 level=INFO source=runner.go:67 msg="discovering available GPUs..."
time=2025-11-19T20:26:35.501+08:00 level=INFO source=server.go:392 msg="starting runner" cmd="C:\Users\qihou\AppData\Local\Programs\Ollama\ollama.exe runner --ollama-engine --port 3846"
time=2025-11-19T20:26:35.733+08:00 level=INFO source=server.go:392 msg="starting runner" cmd="C:\Users\qihou\AppData\Local\Programs\Ollama\ollama.exe runner --ollama-engine --port 3855"
time=2025-11-19T20:26:35.946+08:00 level=INFO source=server.go:392 msg="starting runner" cmd="C:\Users\qihou\AppData\Local\Programs\Ollama\ollama.exe runner --ollama-engine --port 3864"
time=2025-11-19T20:26:36.065+08:00 level=INFO source=runner.go:98 msg="experimental Vulkan support disabled. To enable, set OLLAMA_VULKAN=1"
time=2025-11-19T20:26:36.065+08:00 level=INFO source=types.go:42 msg="inference compute" id=GPU-8b00aeb2-445d-e6d4-fdfe-b7cb31fbf339 filter_id="" library=CUDA compute=12.0 name=CUDA0 description="NVIDIA GeForce RTX 5090" libdirs=ollama,cuda_v12 driver=13.0 pci_id=0000:0e:00.0 type=discrete total="31.8 GiB" available="29.6 GiB"
time=2025-11-19T20:26:46.393+08:00 level=INFO source=server.go:392 msg="starting runner" cmd="C:\Users\qihou\AppData\Local\Programs\Ollama\ollama.exe runner --ollama-engine --port 3896"
time=2025-11-19T20:26:46.613+08:00 level=INFO source=cpu_windows.go:148 msg=packages count=1
time=2025-11-19T20:26:46.613+08:00 level=INFO source=cpu_windows.go:195 msg="" package=0 cores=16 efficiency=0 threads=32
time=2025-11-19T20:26:46.669+08:00 level=INFO source=server.go:209 msg="enabling flash attention"
time=2025-11-19T20:26:46.671+08:00 level=INFO source=server.go:392 msg="starting runner" cmd="C:\Users\qihou\AppData\Local\Programs\Ollama\ollama.exe runner --ollama-engine --model Q:\ollama_models\blobs\sha256-a8cc1361f3145dc01f6d77c6c82c9116b9ffe3c97b34716fe20418455876c40e --port 3905"
time=2025-11-19T20:26:46.676+08:00 level=INFO source=sched.go:443 msg="system memory" total="63.9 GiB" free="46.0 GiB" free_swap="89.8 GiB"
time=2025-11-19T20:26:46.676+08:00 level=INFO source=sched.go:450 msg="gpu memory" id=GPU-8b00aeb2-445d-e6d4-fdfe-b7cb31fbf339 library=CUDA available="29.2 GiB" free="29.6 GiB" minimum="457.0 MiB" overhead="0 B"
time=2025-11-19T20:26:46.677+08:00 level=INFO source=server.go:702 msg="loading model" "model layers"=41 requested=-1
time=2025-11-19T20:26:46.719+08:00 level=INFO source=runner.go:1398 msg="starting ollama engine"
time=2025-11-19T20:26:46.740+08:00 level=INFO source=runner.go:1433 msg="Server listening on 127.0.0.1:3905"
time=2025-11-19T20:26:46.744+08:00 level=INFO source=runner.go:1271 msg=load request="{Operation:fit LoraPath:[] Parallel:2 BatchSize:512 FlashAttention:true KvSize:8192 KvCacheType: NumThreads:16 GPULayers:41[ID:GPU-8b00aeb2-445d-e6d4-fdfe-b7cb31fbf339 Layers:41(0..40)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-11-19T20:26:46.767+08:00 level=INFO source=ggml.go:136 msg="" architecture=qwen3 file_type=Q4_K_M name="Qwen3 14B" description="" num_tensors=443 num_key_values=28
load_backend: loaded CPU backend from C:\Users\qihou\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll
ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 1 CUDA devices:
Device 0: NVIDIA GeForce RTX 5090, compute capability 12.0, VMM: yes, ID: GPU-8b00aeb2-445d-e6d4-fdfe-b7cb31fbf339
load_backend: loaded CUDA backend from C:\Users\qihou\AppData\Local\Programs\Ollama\lib\ollama\cuda_v12\ggml-cuda.dll
time=2025-11-19T20:26:46.880+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=500,520,600,610,700,750,800,860,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(clang)
time=2025-11-19T20:26:47.415+08:00 level=INFO source=runner.go:1271 msg=load request="{Operation:alloc LoraPath:[] Parallel:2 BatchSize:512 FlashAttention:true KvSize:8192 KvCacheType: NumThreads:16 GPULayers:41[ID:GPU-8b00aeb2-445d-e6d4-fdfe-b7cb31fbf339 Layers:41(0..40)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-11-19T20:26:47.568+08:00 level=INFO source=runner.go:1271 msg=load request="{Operation:commit LoraPath:[] Parallel:2 BatchSize:512 FlashAttention:true KvSize:8192 KvCacheType: NumThreads:16 GPULayers:41[ID:GPU-8b00aeb2-445d-e6d4-fdfe-b7cb31fbf339 Layers:41(0..40)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-11-19T20:26:47.568+08:00 level=INFO source=device.go:240 msg="model weights" device=CUDA0 size="8.2 GiB"
time=2025-11-19T20:26:47.568+08:00 level=INFO source=device.go:245 msg="model weights" device=CPU size="417.3 MiB"
time=2025-11-19T20:26:47.569+08:00 level=INFO source=device.go:251 msg="kv cache" device=CUDA0 size="1.2 GiB"
time=2025-11-19T20:26:47.569+08:00 level=INFO source=device.go:262 msg="compute graph" device=CUDA0 size="224.0 MiB"
time=2025-11-19T20:26:47.569+08:00 level=INFO source=device.go:267 msg="compute graph" device=CPU size="10.0 MiB"
time=2025-11-19T20:26:47.569+08:00 level=INFO source=device.go:272 msg="total memory" size="10.1 GiB"
time=2025-11-19T20:26:47.569+08:00 level=INFO source=sched.go:517 msg="loaded runners" count=1
time=2025-11-19T20:26:47.569+08:00 level=INFO source=server.go:1294 msg="waiting for llama runner to start responding"
time=2025-11-19T20:26:47.569+08:00 level=INFO source=ggml.go:482 msg="offloading 40 repeating layers to GPU"
time=2025-11-19T20:26:47.569+08:00 level=INFO source=ggml.go:489 msg="offloading output layer to GPU"
time=2025-11-19T20:26:47.569+08:00 level=INFO source=ggml.go:494 msg="offloaded 41/41 layers to GPU"
time=2025-11-19T20:26:47.569+08:00 level=INFO source=server.go:1328 msg="waiting for server to become available" status="llm server loading model"
time=2025-11-19T20:26:53.577+08:00 level=INFO source=server.go:1332 msg="llama runner started in 6.90 seconds"
[GIN] 2025/11/19 - 20:26:58 | 200 | 11.9365493s | 127.0.0.1 | POST "/api/chat"
[GIN] 2025/11/19 - 20:27:04 | 200 | 18.1134217s | 127.0.0.1 | POST "/api/chat"

<!-- gh-comment-id:3552462625 --> @qihouji commented on GitHub (Nov 19, 2025): Startup log: time=2025-11-19T20:26:35.460+08:00 level=INFO source=routes.go:1544 msg="server config" env="map[CUDA_VISIBLE_DEVICES:0 GGML_VK_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:4096 OLLAMA_DEBUG:INFO OLLAMA_FLASH_ATTENTION:true OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434d OLLAMA_KEEP_ALIVE:2562047h47m16.854775807s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:2 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:Q:\\ollama_models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:2 OLLAMA_ORIGINS:[* http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_REMOTES:[ollama.com] OLLAMA_SCHED_SPREAD:false OLLAMA_VULKAN:false ROCR_VISIBLE_DEVICES:]" time=2025-11-19T20:26:35.477+08:00 level=INFO source=images.go:522 msg="total blobs: 21" time=2025-11-19T20:26:35.478+08:00 level=INFO source=images.go:529 msg="total unused blobs removed: 0" time=2025-11-19T20:26:35.479+08:00 level=INFO source=routes.go:1597 msg="Listening on 127.0.0.1:11434 (version 0.12.11)" time=2025-11-19T20:26:35.481+08:00 level=INFO source=runner.go:67 msg="discovering available GPUs..." time=2025-11-19T20:26:35.501+08:00 level=INFO source=server.go:392 msg="starting runner" cmd="C:\\Users\\qihou\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 3846" time=2025-11-19T20:26:35.733+08:00 level=INFO source=server.go:392 msg="starting runner" cmd="C:\\Users\\qihou\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 3855" time=2025-11-19T20:26:35.946+08:00 level=INFO source=server.go:392 msg="starting runner" cmd="C:\\Users\\qihou\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 3864" time=2025-11-19T20:26:36.065+08:00 level=INFO source=runner.go:98 msg="experimental Vulkan support disabled. To enable, set OLLAMA_VULKAN=1" time=2025-11-19T20:26:36.065+08:00 level=INFO source=types.go:42 msg="inference compute" id=GPU-8b00aeb2-445d-e6d4-fdfe-b7cb31fbf339 filter_id="" library=CUDA compute=12.0 name=CUDA0 description="NVIDIA GeForce RTX 5090" libdirs=ollama,cuda_v12 driver=13.0 pci_id=0000:0e:00.0 type=discrete total="31.8 GiB" available="29.6 GiB" time=2025-11-19T20:26:46.393+08:00 level=INFO source=server.go:392 msg="starting runner" cmd="C:\\Users\\qihou\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 3896" time=2025-11-19T20:26:46.613+08:00 level=INFO source=cpu_windows.go:148 msg=packages count=1 time=2025-11-19T20:26:46.613+08:00 level=INFO source=cpu_windows.go:195 msg="" package=0 cores=16 efficiency=0 threads=32 time=2025-11-19T20:26:46.669+08:00 level=INFO source=server.go:209 msg="enabling flash attention" time=2025-11-19T20:26:46.671+08:00 level=INFO source=server.go:392 msg="starting runner" cmd="C:\\Users\\qihou\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --model Q:\\ollama_models\\blobs\\sha256-a8cc1361f3145dc01f6d77c6c82c9116b9ffe3c97b34716fe20418455876c40e --port 3905" time=2025-11-19T20:26:46.676+08:00 level=INFO source=sched.go:443 msg="system memory" total="63.9 GiB" free="46.0 GiB" free_swap="89.8 GiB" time=2025-11-19T20:26:46.676+08:00 level=INFO source=sched.go:450 msg="gpu memory" id=GPU-8b00aeb2-445d-e6d4-fdfe-b7cb31fbf339 library=CUDA available="29.2 GiB" free="29.6 GiB" minimum="457.0 MiB" overhead="0 B" time=2025-11-19T20:26:46.677+08:00 level=INFO source=server.go:702 msg="loading model" "model layers"=41 requested=-1 time=2025-11-19T20:26:46.719+08:00 level=INFO source=runner.go:1398 msg="starting ollama engine" time=2025-11-19T20:26:46.740+08:00 level=INFO source=runner.go:1433 msg="Server listening on 127.0.0.1:3905" time=2025-11-19T20:26:46.744+08:00 level=INFO source=runner.go:1271 msg=load request="{Operation:fit LoraPath:[] Parallel:2 BatchSize:512 FlashAttention:true KvSize:8192 KvCacheType: NumThreads:16 GPULayers:41[ID:GPU-8b00aeb2-445d-e6d4-fdfe-b7cb31fbf339 Layers:41(0..40)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-11-19T20:26:46.767+08:00 level=INFO source=ggml.go:136 msg="" architecture=qwen3 file_type=Q4_K_M name="Qwen3 14B" description="" num_tensors=443 num_key_values=28 load_backend: loaded CPU backend from C:\Users\qihou\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-haswell.dll ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no ggml_cuda_init: found 1 CUDA devices: Device 0: NVIDIA GeForce RTX 5090, compute capability 12.0, VMM: yes, ID: GPU-8b00aeb2-445d-e6d4-fdfe-b7cb31fbf339 load_backend: loaded CUDA backend from C:\Users\qihou\AppData\Local\Programs\Ollama\lib\ollama\cuda_v12\ggml-cuda.dll time=2025-11-19T20:26:46.880+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=500,520,600,610,700,750,800,860,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(clang) time=2025-11-19T20:26:47.415+08:00 level=INFO source=runner.go:1271 msg=load request="{Operation:alloc LoraPath:[] Parallel:2 BatchSize:512 FlashAttention:true KvSize:8192 KvCacheType: NumThreads:16 GPULayers:41[ID:GPU-8b00aeb2-445d-e6d4-fdfe-b7cb31fbf339 Layers:41(0..40)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-11-19T20:26:47.568+08:00 level=INFO source=runner.go:1271 msg=load request="{Operation:commit LoraPath:[] Parallel:2 BatchSize:512 FlashAttention:true KvSize:8192 KvCacheType: NumThreads:16 GPULayers:41[ID:GPU-8b00aeb2-445d-e6d4-fdfe-b7cb31fbf339 Layers:41(0..40)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-11-19T20:26:47.568+08:00 level=INFO source=device.go:240 msg="model weights" device=CUDA0 size="8.2 GiB" time=2025-11-19T20:26:47.568+08:00 level=INFO source=device.go:245 msg="model weights" device=CPU size="417.3 MiB" time=2025-11-19T20:26:47.569+08:00 level=INFO source=device.go:251 msg="kv cache" device=CUDA0 size="1.2 GiB" time=2025-11-19T20:26:47.569+08:00 level=INFO source=device.go:262 msg="compute graph" device=CUDA0 size="224.0 MiB" time=2025-11-19T20:26:47.569+08:00 level=INFO source=device.go:267 msg="compute graph" device=CPU size="10.0 MiB" time=2025-11-19T20:26:47.569+08:00 level=INFO source=device.go:272 msg="total memory" size="10.1 GiB" time=2025-11-19T20:26:47.569+08:00 level=INFO source=sched.go:517 msg="loaded runners" count=1 time=2025-11-19T20:26:47.569+08:00 level=INFO source=server.go:1294 msg="waiting for llama runner to start responding" time=2025-11-19T20:26:47.569+08:00 level=INFO source=ggml.go:482 msg="offloading 40 repeating layers to GPU" time=2025-11-19T20:26:47.569+08:00 level=INFO source=ggml.go:489 msg="offloading output layer to GPU" time=2025-11-19T20:26:47.569+08:00 level=INFO source=ggml.go:494 msg="offloaded 41/41 layers to GPU" time=2025-11-19T20:26:47.569+08:00 level=INFO source=server.go:1328 msg="waiting for server to become available" status="llm server loading model" time=2025-11-19T20:26:53.577+08:00 level=INFO source=server.go:1332 msg="llama runner started in 6.90 seconds" [GIN] 2025/11/19 - 20:26:58 | 200 | 11.9365493s | 127.0.0.1 | POST "/api/chat" [GIN] 2025/11/19 - 20:27:04 | 200 | 18.1134217s | 127.0.0.1 | POST "/api/chat"
GiteaMirror commented 2026-04-29 08:26:54 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 19, 2025):

Perhaps https://github.com/ollama/ollama/issues/13112, should be fixed in 0.13.0 (currently in pre-release).

<!-- gh-comment-id:3553939116 --> @rick-github commented on GitHub (Nov 19, 2025): Perhaps https://github.com/ollama/ollama/issues/13112, should be fixed in 0.13.0 (currently in [pre-release](https://github.com/ollama/ollama/releases/tag/v0.13.0-rc0)).
GiteaMirror commented 2026-04-29 08:26:55 -05:00
Author
Owner
Copy Link

@qihouji commented on GitHub (Nov 21, 2025):

fixed in 0.13.0, thanks

<!-- gh-comment-id:3560784003 --> @qihouji commented on GitHub (Nov 21, 2025): fixed in 0.13.0, thanks
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
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#55172
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?