github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-08 17:49:24 -05:00
Code Issues 14.6k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #13536] Ollama not using the GPU with NVIDIA driver version 591.59 #70975

New Issue
Closed
opened 2026-05-04 23:37:10 -05:00 by GiteaMirror · 3 comments
GiteaMirror commented 2026-05-04 23:37:10 -05:00
Owner
Copy Link

Originally created by @I-had-a-bad-idea on GitHub (Dec 20, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/13536

What is the issue?

Ollama doesnt seem to use the GPU.
I am on the NVIDIA driver release from the 18th December (Version 591.59).
It falls back to the CPU but is of course a lot slower.

I replaced my name with <my_name>.

PS C:\Users<my_nam> nvidia-smi
Sat Dec 20 18:52:21 2025
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 591.59 Driver Version: 591.59 CUDA Version: 13.1 |
+-----------------------------------------+------------------------+----------------------+
| GPU Name Driver-Model | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+========================+======================|
| 0 NVIDIA GeForce RTX 4050 ... WDDM | 00000000:01:00.0 Off | N/A |
| N/A 46C P8 2W / 30W | 105MiB / 6141MiB | 0% Default |
| | | N/A |
+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=========================================================================================|
| 0 N/A N/A 22568 C ...al\Programs\Ollama\ollama.exe N/A |
+-----------------------------------------------------------------------------------------+

PS C:\Users<my_name> ollama ps
NAME ID SIZE PROCESSOR CONTEXT UNTIL
qwen2.5vl:7b-q4_K_M 5ced39dfa4ba 13 GB 100% CPU 4096 4 minutes from now

Relevant log output

time=2025-12-20T18:35:36.120+01:00 level=INFO source=routes.go:1554 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:INFO OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:C:\\Users\\<my_name>\\.ollama\\models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:1 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_REMOTES:[ollama.com] OLLAMA_SCHED_SPREAD:false OLLAMA_VULKAN:false ROCR_VISIBLE_DEVICES:]"
time=2025-12-20T18:35:36.164+01:00 level=INFO source=images.go:493 msg="total blobs: 15"
time=2025-12-20T18:35:36.166+01:00 level=INFO source=images.go:500 msg="total unused blobs removed: 0"
time=2025-12-20T18:35:36.168+01:00 level=INFO source=routes.go:1607 msg="Listening on 127.0.0.1:11434 (version 0.13.5)"
time=2025-12-20T18:35:36.170+01:00 level=INFO source=runner.go:67 msg="discovering available GPUs..."
time=2025-12-20T18:35:36.195+01:00 level=INFO source=server.go:429 msg="starting runner" cmd="C:\\Users\\<my_name>\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 54509"
time=2025-12-20T18:35:38.650+01:00 level=INFO source=server.go:429 msg="starting runner" cmd="C:\\Users\\<my_name>\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 54519"
time=2025-12-20T18:35:39.726+01:00 level=INFO source=server.go:429 msg="starting runner" cmd="C:\\Users\\<my_name>\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 54533"
time=2025-12-20T18:35:39.955+01:00 level=INFO source=runner.go:106 msg="experimental Vulkan support disabled.  To enable, set OLLAMA_VULKAN=1"
time=2025-12-20T18:35:39.961+01:00 level=INFO source=server.go:429 msg="starting runner" cmd="C:\\Users\\<my_name>\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 54543"
time=2025-12-20T18:35:39.962+01:00 level=INFO source=server.go:429 msg="starting runner" cmd="C:\\Users\\<my_name>\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 54544"
time=2025-12-20T18:35:40.619+01:00 level=INFO source=types.go:42 msg="inference compute" id=GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c filter_id="" library=CUDA compute=8.9 name=CUDA0 description="NVIDIA GeForce RTX 4050 Laptop GPU" libdirs=ollama,cuda_v13 driver=13.1 pci_id=0000:01:00.0 type=discrete total="6.0 GiB" available="5.8 GiB"
time=2025-12-20T18:35:40.619+01:00 level=INFO source=routes.go:1648 msg="entering low vram mode" "total vram"="6.0 GiB" threshold="20.0 GiB"
[GIN] 2025/12/20 - 18:35:40 | 200 |      5.9829ms |       127.0.0.1 | GET      "/api/version"
[GIN] 2025/12/20 - 18:45:36 | 200 |            0s |       127.0.0.1 | GET      "/api/version"
[GIN] 2025/12/20 - 18:45:36 | 200 |     18.8023ms |       127.0.0.1 | GET      "/api/version"
[GIN] 2025/12/20 - 18:45:36 | 200 |     39.3302ms |       127.0.0.1 | GET      "/api/tags"
[GIN] 2025/12/20 - 18:45:37 | 401 |    302.7122ms |       127.0.0.1 | POST     "/api/me"
[GIN] 2025/12/20 - 18:45:37 | 401 |    272.3223ms |       127.0.0.1 | POST     "/api/me"
[GIN] 2025/12/20 - 18:45:37 | 200 |    136.2788ms |       127.0.0.1 | POST     "/api/show"
[GIN] 2025/12/20 - 18:45:43 | 200 |     46.4007ms |       127.0.0.1 | POST     "/api/show"
[GIN] 2025/12/20 - 18:45:45 | 200 |      1.5518ms |       127.0.0.1 | GET      "/api/tags"
[GIN] 2025/12/20 - 18:45:45 | 200 |      61.572ms |       127.0.0.1 | POST     "/api/show"
[GIN] 2025/12/20 - 18:45:45 | 200 |     35.6559ms |       127.0.0.1 | POST     "/api/show"
time=2025-12-20T18:45:45.494+01:00 level=INFO source=server.go:429 msg="starting runner" cmd="C:\\Users\\<my_name>\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 49864"
time=2025-12-20T18:45:47.298+01:00 level=INFO source=cpu_windows.go:148 msg=packages count=1
time=2025-12-20T18:45:47.298+01:00 level=INFO source=cpu_windows.go:164 msg="efficiency cores detected" maxEfficiencyClass=1
time=2025-12-20T18:45:47.298+01:00 level=INFO source=cpu_windows.go:195 msg="" package=0 cores=8 efficiency=4 threads=12
time=2025-12-20T18:45:47.353+01:00 level=INFO source=server.go:429 msg="starting runner" cmd="C:\\Users\\<my_name>\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --model C:\\Users\\<my_name>\\.ollama\\models\\blobs\\sha256-a99b7f834d754b88f122d865f32758ba9f0994a83f8363df2c1e71c17605a025 --port 49874"
time=2025-12-20T18:45:47.368+01:00 level=INFO source=sched.go:443 msg="system memory" total="31.7 GiB" free="17.8 GiB" free_swap="21.7 GiB"
time=2025-12-20T18:45:47.368+01:00 level=INFO source=sched.go:450 msg="gpu memory" id=GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c library=CUDA available="5.3 GiB" free="5.8 GiB" minimum="457.0 MiB" overhead="0 B"
time=2025-12-20T18:45:47.368+01:00 level=INFO source=server.go:746 msg="loading model" "model layers"=29 requested=-1
time=2025-12-20T18:45:47.438+01:00 level=INFO source=runner.go:1405 msg="starting ollama engine"
time=2025-12-20T18:45:47.470+01:00 level=INFO source=runner.go:1440 msg="Server listening on 127.0.0.1:49874"
time=2025-12-20T18:45:47.478+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:29[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:29(0..28)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:45:47.500+01:00 level=INFO source=ggml.go:136 msg="" architecture=qwen25vl file_type=Q4_K_M name="" description="" num_tensors=858 num_key_values=36
load_backend: loaded CPU backend from C:\Users\<my_name>\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-alderlake.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 4050 Laptop GPU, compute capability 8.9, VMM: yes, ID: GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c
load_backend: loaded CUDA backend from C:\Users\<my_name>\AppData\Local\Programs\Ollama\lib\ollama\cuda_v13\ggml-cuda.dll
time=2025-12-20T18:45:48.174+01:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX_VNNI=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=750,800,860,870,890,900,1000,1030,1100,1200,1210 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(clang)
time=2025-12-20T18:45:49.603+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:45:49.999+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:27[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:27(1..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:45:50.382+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:26[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:26(2..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:45:50.774+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:25[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:25(3..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:45:51.129+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:24[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:24(4..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:45:51.479+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:23[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:23(5..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:45:51.854+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:22[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:22(6..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:45:52.214+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:21[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:21(7..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:45:52.566+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:20[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:20(8..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:45:52.951+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:19[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:19(9..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:45:53.337+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:18[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:18(10..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:45:53.712+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:17[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:17(11..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:45:54.086+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:16[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:16(12..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:45:54.464+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:15[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:15(13..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:45:54.862+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:14[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:14(14..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:45:55.414+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:13[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:13(15..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:45:56.229+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:12[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:12(16..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:45:58.027+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:11[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:11(17..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:45:58.971+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:10[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:10(18..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:45:59.378+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:9[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:9(19..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:45:59.764+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:8[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:8(20..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:00.148+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:7[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:7(21..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:00.527+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:6[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:6(22..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:00.910+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:5[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:5(23..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:01.291+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:4[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:4(24..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:01.697+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:3[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:3(25..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:02.069+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:2[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:2(26..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:02.458+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:1[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:1(27..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:02.830+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:03.213+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:03.779+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:27[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:27(1..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:05.215+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:26[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:26(2..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:07.010+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:25[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:25(3..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
[GIN] 2025/12/20 - 18:46:08 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2025/12/20 - 18:46:08 | 200 |            0s |       127.0.0.1 | GET      "/api/ps"
time=2025-12-20T18:46:08.623+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:24[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:24(4..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:10.450+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:23[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:23(5..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:12.114+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:22[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:22(6..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:13.649+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:21[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:21(7..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:15.164+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:20[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:20(8..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
[GIN] 2025/12/20 - 18:46:15 | 200 |      2.1197ms |       127.0.0.1 | GET      "/api/tags"
time=2025-12-20T18:46:16.682+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:19[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:19(9..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:18.193+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:18[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:18(10..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:19.663+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:17[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:17(11..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:21.066+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:16[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:16(12..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:22.426+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:15[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:15(13..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:23.808+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:14[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:14(14..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:25.199+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:13[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:13(15..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:26.504+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:12[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:12(16..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:27.810+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:11[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:11(17..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:29.039+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:10[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:10(18..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:30.260+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:9[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:9(19..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:31.426+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:8[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:8(20..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:32.629+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:7[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:7(21..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:33.746+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:6[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:6(22..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:34.896+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:5[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:5(23..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:36.204+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:4[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:4(24..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:37.280+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:3[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:3(25..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:38.319+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:2[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:2(26..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:39.352+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:1[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:1(27..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:40.363+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:41.136+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-12-20T18:46:41.136+01:00 level=INFO source=ggml.go:482 msg="offloading 0 repeating layers to GPU"
time=2025-12-20T18:46:41.136+01:00 level=INFO source=ggml.go:486 msg="offloading output layer to CPU"
time=2025-12-20T18:46:41.136+01:00 level=INFO source=ggml.go:494 msg="offloaded 0/29 layers to GPU"
time=2025-12-20T18:46:41.137+01:00 level=INFO source=device.go:245 msg="model weights" device=CPU size="5.6 GiB"
time=2025-12-20T18:46:41.137+01:00 level=INFO source=device.go:256 msg="kv cache" device=CPU size="224.0 MiB"
time=2025-12-20T18:46:41.137+01:00 level=INFO source=device.go:267 msg="compute graph" device=CPU size="6.7 GiB"
time=2025-12-20T18:46:41.137+01:00 level=INFO source=device.go:272 msg="total memory" size="12.5 GiB"
time=2025-12-20T18:46:41.137+01:00 level=INFO source=sched.go:517 msg="loaded runners" count=1
time=2025-12-20T18:46:41.137+01:00 level=INFO source=server.go:1338 msg="waiting for llama runner to start responding"
time=2025-12-20T18:46:41.137+01:00 level=INFO source=server.go:1372 msg="waiting for server to become available" status="llm server loading model"
[GIN] 2025/12/20 - 18:46:45 | 200 |      2.6683ms |       127.0.0.1 | GET      "/api/tags"
time=2025-12-20T18:46:47.901+01:00 level=INFO source=server.go:1376 msg="llama runner started in 60.53 seconds"
[GIN] 2025/12/20 - 18:46:50 | 200 |          1m4s |       127.0.0.1 | POST     "/api/chat"

OS

Windows

GPU

Nvidia

CPU

Intel

Ollama version

0.13.5

Originally created by @I-had-a-bad-idea on GitHub (Dec 20, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/13536 ### What is the issue? Ollama doesnt seem to use the GPU. I am on the NVIDIA driver release from the 18th December (Version 591.59). It falls back to the CPU but is of course a lot slower. I replaced my name with <my_name>. PS C:\Users\<my_nam> nvidia-smi Sat Dec 20 18:52:21 2025 +-----------------------------------------------------------------------------------------+ | NVIDIA-SMI 591.59 Driver Version: 591.59 CUDA Version: 13.1 | +-----------------------------------------+------------------------+----------------------+ | GPU Name Driver-Model | Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. | | | | MIG M. | |=========================================+========================+======================| | 0 NVIDIA GeForce RTX 4050 ... WDDM | 00000000:01:00.0 Off | N/A | | N/A 46C P8 2W / 30W | 105MiB / 6141MiB | 0% Default | | | | N/A | +-----------------------------------------+------------------------+----------------------+ +-----------------------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=========================================================================================| | 0 N/A N/A 22568 C ...al\Programs\Ollama\ollama.exe N/A | +-----------------------------------------------------------------------------------------+ PS C:\Users\<my_name> ollama ps NAME ID SIZE PROCESSOR CONTEXT UNTIL qwen2.5vl:7b-q4_K_M 5ced39dfa4ba 13 GB 100% CPU 4096 4 minutes from now ### Relevant log output ```shell time=2025-12-20T18:35:36.120+01:00 level=INFO source=routes.go:1554 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:INFO OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:C:\\Users\\<my_name>\\.ollama\\models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:1 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_REMOTES:[ollama.com] OLLAMA_SCHED_SPREAD:false OLLAMA_VULKAN:false ROCR_VISIBLE_DEVICES:]" time=2025-12-20T18:35:36.164+01:00 level=INFO source=images.go:493 msg="total blobs: 15" time=2025-12-20T18:35:36.166+01:00 level=INFO source=images.go:500 msg="total unused blobs removed: 0" time=2025-12-20T18:35:36.168+01:00 level=INFO source=routes.go:1607 msg="Listening on 127.0.0.1:11434 (version 0.13.5)" time=2025-12-20T18:35:36.170+01:00 level=INFO source=runner.go:67 msg="discovering available GPUs..." time=2025-12-20T18:35:36.195+01:00 level=INFO source=server.go:429 msg="starting runner" cmd="C:\\Users\\<my_name>\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 54509" time=2025-12-20T18:35:38.650+01:00 level=INFO source=server.go:429 msg="starting runner" cmd="C:\\Users\\<my_name>\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 54519" time=2025-12-20T18:35:39.726+01:00 level=INFO source=server.go:429 msg="starting runner" cmd="C:\\Users\\<my_name>\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 54533" time=2025-12-20T18:35:39.955+01:00 level=INFO source=runner.go:106 msg="experimental Vulkan support disabled. To enable, set OLLAMA_VULKAN=1" time=2025-12-20T18:35:39.961+01:00 level=INFO source=server.go:429 msg="starting runner" cmd="C:\\Users\\<my_name>\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 54543" time=2025-12-20T18:35:39.962+01:00 level=INFO source=server.go:429 msg="starting runner" cmd="C:\\Users\\<my_name>\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 54544" time=2025-12-20T18:35:40.619+01:00 level=INFO source=types.go:42 msg="inference compute" id=GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c filter_id="" library=CUDA compute=8.9 name=CUDA0 description="NVIDIA GeForce RTX 4050 Laptop GPU" libdirs=ollama,cuda_v13 driver=13.1 pci_id=0000:01:00.0 type=discrete total="6.0 GiB" available="5.8 GiB" time=2025-12-20T18:35:40.619+01:00 level=INFO source=routes.go:1648 msg="entering low vram mode" "total vram"="6.0 GiB" threshold="20.0 GiB" [GIN] 2025/12/20 - 18:35:40 | 200 | 5.9829ms | 127.0.0.1 | GET "/api/version" [GIN] 2025/12/20 - 18:45:36 | 200 | 0s | 127.0.0.1 | GET "/api/version" [GIN] 2025/12/20 - 18:45:36 | 200 | 18.8023ms | 127.0.0.1 | GET "/api/version" [GIN] 2025/12/20 - 18:45:36 | 200 | 39.3302ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/12/20 - 18:45:37 | 401 | 302.7122ms | 127.0.0.1 | POST "/api/me" [GIN] 2025/12/20 - 18:45:37 | 401 | 272.3223ms | 127.0.0.1 | POST "/api/me" [GIN] 2025/12/20 - 18:45:37 | 200 | 136.2788ms | 127.0.0.1 | POST "/api/show" [GIN] 2025/12/20 - 18:45:43 | 200 | 46.4007ms | 127.0.0.1 | POST "/api/show" [GIN] 2025/12/20 - 18:45:45 | 200 | 1.5518ms | 127.0.0.1 | GET "/api/tags" [GIN] 2025/12/20 - 18:45:45 | 200 | 61.572ms | 127.0.0.1 | POST "/api/show" [GIN] 2025/12/20 - 18:45:45 | 200 | 35.6559ms | 127.0.0.1 | POST "/api/show" time=2025-12-20T18:45:45.494+01:00 level=INFO source=server.go:429 msg="starting runner" cmd="C:\\Users\\<my_name>\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 49864" time=2025-12-20T18:45:47.298+01:00 level=INFO source=cpu_windows.go:148 msg=packages count=1 time=2025-12-20T18:45:47.298+01:00 level=INFO source=cpu_windows.go:164 msg="efficiency cores detected" maxEfficiencyClass=1 time=2025-12-20T18:45:47.298+01:00 level=INFO source=cpu_windows.go:195 msg="" package=0 cores=8 efficiency=4 threads=12 time=2025-12-20T18:45:47.353+01:00 level=INFO source=server.go:429 msg="starting runner" cmd="C:\\Users\\<my_name>\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --model C:\\Users\\<my_name>\\.ollama\\models\\blobs\\sha256-a99b7f834d754b88f122d865f32758ba9f0994a83f8363df2c1e71c17605a025 --port 49874" time=2025-12-20T18:45:47.368+01:00 level=INFO source=sched.go:443 msg="system memory" total="31.7 GiB" free="17.8 GiB" free_swap="21.7 GiB" time=2025-12-20T18:45:47.368+01:00 level=INFO source=sched.go:450 msg="gpu memory" id=GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c library=CUDA available="5.3 GiB" free="5.8 GiB" minimum="457.0 MiB" overhead="0 B" time=2025-12-20T18:45:47.368+01:00 level=INFO source=server.go:746 msg="loading model" "model layers"=29 requested=-1 time=2025-12-20T18:45:47.438+01:00 level=INFO source=runner.go:1405 msg="starting ollama engine" time=2025-12-20T18:45:47.470+01:00 level=INFO source=runner.go:1440 msg="Server listening on 127.0.0.1:49874" time=2025-12-20T18:45:47.478+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:29[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:29(0..28)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:45:47.500+01:00 level=INFO source=ggml.go:136 msg="" architecture=qwen25vl file_type=Q4_K_M name="" description="" num_tensors=858 num_key_values=36 load_backend: loaded CPU backend from C:\Users\<my_name>\AppData\Local\Programs\Ollama\lib\ollama\ggml-cpu-alderlake.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 4050 Laptop GPU, compute capability 8.9, VMM: yes, ID: GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c load_backend: loaded CUDA backend from C:\Users\<my_name>\AppData\Local\Programs\Ollama\lib\ollama\cuda_v13\ggml-cuda.dll time=2025-12-20T18:45:48.174+01:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX_VNNI=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=750,800,860,870,890,900,1000,1030,1100,1200,1210 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(clang) time=2025-12-20T18:45:49.603+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:45:49.999+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:27[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:27(1..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:45:50.382+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:26[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:26(2..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:45:50.774+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:25[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:25(3..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:45:51.129+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:24[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:24(4..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:45:51.479+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:23[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:23(5..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:45:51.854+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:22[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:22(6..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:45:52.214+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:21[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:21(7..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:45:52.566+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:20[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:20(8..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:45:52.951+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:19[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:19(9..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:45:53.337+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:18[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:18(10..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:45:53.712+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:17[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:17(11..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:45:54.086+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:16[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:16(12..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:45:54.464+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:15[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:15(13..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:45:54.862+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:14[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:14(14..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:45:55.414+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:13[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:13(15..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:45:56.229+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:12[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:12(16..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:45:58.027+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:11[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:11(17..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:45:58.971+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:10[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:10(18..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:45:59.378+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:9[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:9(19..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:45:59.764+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:8[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:8(20..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:00.148+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:7[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:7(21..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:00.527+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:6[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:6(22..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:00.910+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:5[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:5(23..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:01.291+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:4[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:4(24..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:01.697+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:3[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:3(25..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:02.069+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:2[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:2(26..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:02.458+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:1[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:1(27..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:02.830+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:03.213+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:03.779+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:27[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:27(1..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:05.215+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:26[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:26(2..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:07.010+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:25[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:25(3..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" [GIN] 2025/12/20 - 18:46:08 | 200 | 0s | 127.0.0.1 | HEAD "/" [GIN] 2025/12/20 - 18:46:08 | 200 | 0s | 127.0.0.1 | GET "/api/ps" time=2025-12-20T18:46:08.623+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:24[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:24(4..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:10.450+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:23[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:23(5..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:12.114+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:22[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:22(6..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:13.649+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:21[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:21(7..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:15.164+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:20[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:20(8..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" [GIN] 2025/12/20 - 18:46:15 | 200 | 2.1197ms | 127.0.0.1 | GET "/api/tags" time=2025-12-20T18:46:16.682+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:19[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:19(9..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:18.193+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:18[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:18(10..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:19.663+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:17[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:17(11..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:21.066+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:16[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:16(12..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:22.426+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:15[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:15(13..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:23.808+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:14[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:14(14..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:25.199+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:13[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:13(15..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:26.504+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:12[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:12(16..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:27.810+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:11[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:11(17..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:29.039+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:10[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:10(18..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:30.260+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:9[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:9(19..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:31.426+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:8[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:8(20..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:32.629+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:7[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:7(21..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:33.746+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:6[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:6(22..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:34.896+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:5[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:5(23..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:36.204+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:4[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:4(24..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:37.280+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:3[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:3(25..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:38.319+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:2[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:2(26..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:39.352+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:1[ID:GPU-7f57fbdd-bdf5-3d09-918d-e8a06bce4a1c Layers:1(27..27)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:40.363+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:41.136+01:00 level=INFO source=runner.go:1278 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Disabled KvSize:4096 KvCacheType: NumThreads:4 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-12-20T18:46:41.136+01:00 level=INFO source=ggml.go:482 msg="offloading 0 repeating layers to GPU" time=2025-12-20T18:46:41.136+01:00 level=INFO source=ggml.go:486 msg="offloading output layer to CPU" time=2025-12-20T18:46:41.136+01:00 level=INFO source=ggml.go:494 msg="offloaded 0/29 layers to GPU" time=2025-12-20T18:46:41.137+01:00 level=INFO source=device.go:245 msg="model weights" device=CPU size="5.6 GiB" time=2025-12-20T18:46:41.137+01:00 level=INFO source=device.go:256 msg="kv cache" device=CPU size="224.0 MiB" time=2025-12-20T18:46:41.137+01:00 level=INFO source=device.go:267 msg="compute graph" device=CPU size="6.7 GiB" time=2025-12-20T18:46:41.137+01:00 level=INFO source=device.go:272 msg="total memory" size="12.5 GiB" time=2025-12-20T18:46:41.137+01:00 level=INFO source=sched.go:517 msg="loaded runners" count=1 time=2025-12-20T18:46:41.137+01:00 level=INFO source=server.go:1338 msg="waiting for llama runner to start responding" time=2025-12-20T18:46:41.137+01:00 level=INFO source=server.go:1372 msg="waiting for server to become available" status="llm server loading model" [GIN] 2025/12/20 - 18:46:45 | 200 | 2.6683ms | 127.0.0.1 | GET "/api/tags" time=2025-12-20T18:46:47.901+01:00 level=INFO source=server.go:1376 msg="llama runner started in 60.53 seconds" [GIN] 2025/12/20 - 18:46:50 | 200 | 1m4s | 127.0.0.1 | POST "/api/chat" ``` ### OS Windows ### GPU Nvidia ### CPU Intel ### Ollama version 0.13.5
GiteaMirror added the bug label 2026-05-04 23:37:10 -05:00
GiteaMirror commented 2026-05-04 23:37:14 -05:00
Author
Owner
Copy Link

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

Try setting OLLAMA_LLM_LIBRARY=cuda_v12 in the server environment.

<!-- gh-comment-id:3678228426 --> @rick-github commented on GitHub (Dec 21, 2025): Try setting `OLLAMA_LLM_LIBRARY=cuda_v12` in the [server environment](https://docs.ollama.com/faq#how-do-i-configure-ollama-server).
GiteaMirror commented 2026-05-04 23:37:15 -05:00
Author
Owner
Copy Link

@I-had-a-bad-idea commented on GitHub (Dec 21, 2025):

Sadly didnt help. The problem seems to be the VRAM.
It seems to detect the GPU and try to use it, but then decides it has too little VRAM.
After that it tries to offload less and less layers to the GPU (see log), until it completely runs on the CPU.
I did not have this problem before (last time I know it worked was the 17th of December).
Maybe there were some changes to the memory allocation, which caused this.

<!-- gh-comment-id:3678742042 --> @I-had-a-bad-idea commented on GitHub (Dec 21, 2025): Sadly didnt help. The problem seems to be the VRAM. It seems to detect the GPU and try to use it, but then decides it has too little VRAM. After that it tries to offload less and less layers to the GPU (see log), until it completely runs on the CPU. I did not have this problem before (last time I know it worked was the 17th of December). Maybe there were some changes to the memory allocation, which caused this.
GiteaMirror commented 2026-05-04 23:37:16 -05:00
Author
Owner
Copy Link

@I-had-a-bad-idea commented on GitHub (Dec 21, 2025):

It seems to a problem of the model. Instead of 6 it needs 13 GB VRAM.
Other models work just fine.

<!-- gh-comment-id:3678782830 --> @I-had-a-bad-idea commented on GitHub (Dec 21, 2025): It seems to a problem of the model. Instead of 6 it needs 13 GB VRAM. Other models work just fine.
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#70975
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?