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 #13930] [BUG] Ollama uses CPU instead of GPU #55628

New Issue
Closed
opened 2026-04-29 09:30:18 -05:00 by GiteaMirror · 3 comments
GiteaMirror commented 2026-04-29 09:30:18 -05:00
Owner
Copy Link

Originally created by @discostur on GitHub (Jan 27, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/13930

What is the issue?

Ollama detects my Tesla T4 but uses CPU instead of GPU.

Relevant log output

time=2026-01-27T09:26:19.319Z level=INFO source=routes.go:1631 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:8192 OLLAMA_DEBUG:INFO OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://0.0.0.0:11434 OLLAMA_KEEP_ALIVE:2562047h47m16.854775807s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/root/.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: http_proxy: https_proxy: no_proxy:]"
time=2026-01-27T09:26:19.343Z level=INFO source=images.go:473 msg="total blobs: 56"
time=2026-01-27T09:26:19.345Z level=INFO source=images.go:480 msg="total unused blobs removed: 0"
time=2026-01-27T09:26:19.346Z level=INFO source=routes.go:1684 msg="Listening on [::]:11434 (version 0.15.2)"
time=2026-01-27T09:26:19.349Z level=INFO source=runner.go:67 msg="discovering available GPUs..."
time=2026-01-27T09:26:19.353Z level=INFO source=server.go:429 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --port 40685"
time=2026-01-27T09:26:20.119Z level=INFO source=server.go:429 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --port 46089"
time=2026-01-27T09:26:20.727Z level=INFO source=runner.go:106 msg="experimental Vulkan support disabled.  To enable, set OLLAMA_VULKAN=1"
time=2026-01-27T09:26:20.727Z level=INFO source=server.go:429 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --port 41937"
time=2026-01-27T09:26:20.727Z level=INFO source=server.go:429 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --port 38685"
time=2026-01-27T09:26:20.994Z level=INFO source=types.go:42 msg="inference compute" id=GPU-60d0c2c0-f953-d178-82f0-754b2f08821b filter_id="" library=CUDA compute=7.5 name=CUDA0 description="Tesla T4" libdirs=ollama,cuda_v13 driver=13.1 pci_id=0000:01:00.0 type=discrete total="16.0 GiB" available="15.6 GiB"
time=2026-01-27T09:26:20.994Z level=INFO source=routes.go:1725 msg="entering low vram mode" "total vram"="16.0 GiB" threshold="20.0 GiB"
[GIN] 2026/01/27 - 09:32:39 | 200 |    9.428926ms |     172.18.0.17 | GET      "/api/tags"
[GIN] 2026/01/27 - 09:32:39 | 200 |     122.215µs |     172.18.0.17 | GET      "/api/ps"
[GIN] 2026/01/27 - 09:32:53 | 200 |    4.137045ms |     172.18.0.17 | GET      "/api/tags"
[GIN] 2026/01/27 - 09:32:53 | 200 |      49.385µs |     172.18.0.17 | GET      "/api/ps"
[GIN] 2026/01/27 - 09:32:53 | 200 |      64.695µs |     172.18.0.17 | GET      "/api/version"
time=2026-01-27T09:33:04.803Z level=INFO source=server.go:429 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --port 44837"
time=2026-01-27T09:33:05.003Z level=WARN source=runner.go:356 msg="unable to refresh free memory, using old values"
time=2026-01-27T09:33:05.006Z level=WARN source=cpu_linux.go:130 msg="failed to parse CPU allowed micro secs" error="strconv.ParseInt: parsing \"max\": invalid syntax"
time=2026-01-27T09:33:05.539Z level=INFO source=server.go:245 msg="enabling flash attention"
time=2026-01-27T09:33:05.539Z level=INFO source=server.go:429 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --model /root/.ollama/models/blobs/sha256-10fe673de12c20b74b8d670a9fdf0fd36b43b0a86ffc04daeb175c0a2b98c4f9 --port 34369"
time=2026-01-27T09:33:05.540Z level=INFO source=sched.go:452 msg="system memory" total="125.8 GiB" free="125.5 GiB" free_swap="0 B"
time=2026-01-27T09:33:05.540Z level=INFO source=sched.go:459 msg="gpu memory" id=GPU-60d0c2c0-f953-d178-82f0-754b2f08821b library=CUDA available="15.1 GiB" free="15.6 GiB" minimum="457.0 MiB" overhead="0 B"
time=2026-01-27T09:33:05.540Z level=INFO source=server.go:755 msg="loading model" "model layers"=25 requested=-1
time=2026-01-27T09:33:05.565Z level=INFO source=runner.go:1405 msg="starting ollama engine"
time=2026-01-27T09:33:05.566Z level=INFO source=runner.go:1440 msg="Server listening on 127.0.0.1:34369"
time=2026-01-27T09:33:05.573Z level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:4096 KvCacheType: NumThreads:16 GPULayers:25[ID:GPU-60d0c2c0-f953-d178-82f0-754b2f08821b Layers:25(0..24)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2026-01-27T09:33:05.750Z level=INFO source=ggml.go:136 msg="" architecture=gpt-oss file_type=Q4_K_M name=Gpt-Oss-20B description="" num_tensors=459 num_key_values=38
load_backend: loaded CPU backend from /usr/lib/ollama/libggml-cpu-skylakex.so
ggml_cuda_init: failed to initialize CUDA: no CUDA-capable device is detected
load_backend: loaded CUDA backend from /usr/lib/ollama/cuda_v13/libggml-cuda.so
time=2026-01-27T09:33:05.883Z 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.LLAMAFILE=1 CPU.1.LLAMAFILE=1 compiler=cgo(gcc)
time=2026-01-27T09:33:05.938Z level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:4096 KvCacheType: NumThreads:16 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2026-01-27T09:33:06.161Z level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:4096 KvCacheType: NumThreads:16 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2026-01-27T09:33:06.578Z level=INFO source=runner.go:1278 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:4096 KvCacheType: NumThreads:16 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2026-01-27T09:33:06.578Z level=INFO source=ggml.go:482 msg="offloading 0 repeating layers to GPU"
time=2026-01-27T09:33:06.578Z level=INFO source=ggml.go:486 msg="offloading output layer to CPU"
time=2026-01-27T09:33:06.578Z level=INFO source=ggml.go:494 msg="offloaded 0/25 layers to GPU"
time=2026-01-27T09:33:06.578Z level=INFO source=device.go:245 msg="model weights" device=CPU size="11.0 GiB"
time=2026-01-27T09:33:06.578Z level=INFO source=device.go:256 msg="kv cache" device=CPU size="192.0 MiB"
time=2026-01-27T09:33:06.578Z level=INFO source=device.go:267 msg="compute graph" device=CPU size="94.8 MiB"
time=2026-01-27T09:33:06.578Z level=INFO source=device.go:272 msg="total memory" size="11.3 GiB"
time=2026-01-27T09:33:06.578Z level=INFO source=sched.go:526 msg="loaded runners" count=1
time=2026-01-27T09:33:06.578Z level=INFO source=server.go:1347 msg="waiting for llama runner to start responding"
time=2026-01-27T09:33:06.579Z level=INFO source=server.go:1381 msg="waiting for server to become available" status="llm server loading model"
time=2026-01-27T09:33:14.613Z level=INFO source=server.go:1385 msg="llama runner started in 9.07 seconds"

OS

Linux

GPU

Nvidia

CPU

Intel

Ollama version

0.15.2

Originally created by @discostur on GitHub (Jan 27, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/13930 ### What is the issue? Ollama detects my Tesla T4 but uses CPU instead of GPU. ### Relevant log output ```shell time=2026-01-27T09:26:19.319Z level=INFO source=routes.go:1631 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:8192 OLLAMA_DEBUG:INFO OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://0.0.0.0:11434 OLLAMA_KEEP_ALIVE:2562047h47m16.854775807s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/root/.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: http_proxy: https_proxy: no_proxy:]" time=2026-01-27T09:26:19.343Z level=INFO source=images.go:473 msg="total blobs: 56" time=2026-01-27T09:26:19.345Z level=INFO source=images.go:480 msg="total unused blobs removed: 0" time=2026-01-27T09:26:19.346Z level=INFO source=routes.go:1684 msg="Listening on [::]:11434 (version 0.15.2)" time=2026-01-27T09:26:19.349Z level=INFO source=runner.go:67 msg="discovering available GPUs..." time=2026-01-27T09:26:19.353Z level=INFO source=server.go:429 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --port 40685" time=2026-01-27T09:26:20.119Z level=INFO source=server.go:429 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --port 46089" time=2026-01-27T09:26:20.727Z level=INFO source=runner.go:106 msg="experimental Vulkan support disabled. To enable, set OLLAMA_VULKAN=1" time=2026-01-27T09:26:20.727Z level=INFO source=server.go:429 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --port 41937" time=2026-01-27T09:26:20.727Z level=INFO source=server.go:429 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --port 38685" time=2026-01-27T09:26:20.994Z level=INFO source=types.go:42 msg="inference compute" id=GPU-60d0c2c0-f953-d178-82f0-754b2f08821b filter_id="" library=CUDA compute=7.5 name=CUDA0 description="Tesla T4" libdirs=ollama,cuda_v13 driver=13.1 pci_id=0000:01:00.0 type=discrete total="16.0 GiB" available="15.6 GiB" time=2026-01-27T09:26:20.994Z level=INFO source=routes.go:1725 msg="entering low vram mode" "total vram"="16.0 GiB" threshold="20.0 GiB" [GIN] 2026/01/27 - 09:32:39 | 200 | 9.428926ms | 172.18.0.17 | GET "/api/tags" [GIN] 2026/01/27 - 09:32:39 | 200 | 122.215µs | 172.18.0.17 | GET "/api/ps" [GIN] 2026/01/27 - 09:32:53 | 200 | 4.137045ms | 172.18.0.17 | GET "/api/tags" [GIN] 2026/01/27 - 09:32:53 | 200 | 49.385µs | 172.18.0.17 | GET "/api/ps" [GIN] 2026/01/27 - 09:32:53 | 200 | 64.695µs | 172.18.0.17 | GET "/api/version" time=2026-01-27T09:33:04.803Z level=INFO source=server.go:429 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --port 44837" time=2026-01-27T09:33:05.003Z level=WARN source=runner.go:356 msg="unable to refresh free memory, using old values" time=2026-01-27T09:33:05.006Z level=WARN source=cpu_linux.go:130 msg="failed to parse CPU allowed micro secs" error="strconv.ParseInt: parsing \"max\": invalid syntax" time=2026-01-27T09:33:05.539Z level=INFO source=server.go:245 msg="enabling flash attention" time=2026-01-27T09:33:05.539Z level=INFO source=server.go:429 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --model /root/.ollama/models/blobs/sha256-10fe673de12c20b74b8d670a9fdf0fd36b43b0a86ffc04daeb175c0a2b98c4f9 --port 34369" time=2026-01-27T09:33:05.540Z level=INFO source=sched.go:452 msg="system memory" total="125.8 GiB" free="125.5 GiB" free_swap="0 B" time=2026-01-27T09:33:05.540Z level=INFO source=sched.go:459 msg="gpu memory" id=GPU-60d0c2c0-f953-d178-82f0-754b2f08821b library=CUDA available="15.1 GiB" free="15.6 GiB" minimum="457.0 MiB" overhead="0 B" time=2026-01-27T09:33:05.540Z level=INFO source=server.go:755 msg="loading model" "model layers"=25 requested=-1 time=2026-01-27T09:33:05.565Z level=INFO source=runner.go:1405 msg="starting ollama engine" time=2026-01-27T09:33:05.566Z level=INFO source=runner.go:1440 msg="Server listening on 127.0.0.1:34369" time=2026-01-27T09:33:05.573Z level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:4096 KvCacheType: NumThreads:16 GPULayers:25[ID:GPU-60d0c2c0-f953-d178-82f0-754b2f08821b Layers:25(0..24)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2026-01-27T09:33:05.750Z level=INFO source=ggml.go:136 msg="" architecture=gpt-oss file_type=Q4_K_M name=Gpt-Oss-20B description="" num_tensors=459 num_key_values=38 load_backend: loaded CPU backend from /usr/lib/ollama/libggml-cpu-skylakex.so ggml_cuda_init: failed to initialize CUDA: no CUDA-capable device is detected load_backend: loaded CUDA backend from /usr/lib/ollama/cuda_v13/libggml-cuda.so time=2026-01-27T09:33:05.883Z 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.LLAMAFILE=1 CPU.1.LLAMAFILE=1 compiler=cgo(gcc) time=2026-01-27T09:33:05.938Z level=INFO source=runner.go:1278 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:4096 KvCacheType: NumThreads:16 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2026-01-27T09:33:06.161Z level=INFO source=runner.go:1278 msg=load request="{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:4096 KvCacheType: NumThreads:16 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2026-01-27T09:33:06.578Z level=INFO source=runner.go:1278 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:4096 KvCacheType: NumThreads:16 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2026-01-27T09:33:06.578Z level=INFO source=ggml.go:482 msg="offloading 0 repeating layers to GPU" time=2026-01-27T09:33:06.578Z level=INFO source=ggml.go:486 msg="offloading output layer to CPU" time=2026-01-27T09:33:06.578Z level=INFO source=ggml.go:494 msg="offloaded 0/25 layers to GPU" time=2026-01-27T09:33:06.578Z level=INFO source=device.go:245 msg="model weights" device=CPU size="11.0 GiB" time=2026-01-27T09:33:06.578Z level=INFO source=device.go:256 msg="kv cache" device=CPU size="192.0 MiB" time=2026-01-27T09:33:06.578Z level=INFO source=device.go:267 msg="compute graph" device=CPU size="94.8 MiB" time=2026-01-27T09:33:06.578Z level=INFO source=device.go:272 msg="total memory" size="11.3 GiB" time=2026-01-27T09:33:06.578Z level=INFO source=sched.go:526 msg="loaded runners" count=1 time=2026-01-27T09:33:06.578Z level=INFO source=server.go:1347 msg="waiting for llama runner to start responding" time=2026-01-27T09:33:06.579Z level=INFO source=server.go:1381 msg="waiting for server to become available" status="llm server loading model" time=2026-01-27T09:33:14.613Z level=INFO source=server.go:1385 msg="llama runner started in 9.07 seconds" ``` ### OS Linux ### GPU Nvidia ### CPU Intel ### Ollama version 0.15.2
GiteaMirror added the bug label 2026-04-29 09:30:18 -05:00
GiteaMirror commented 2026-04-29 09:30:21 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jan 27, 2026):

It looks like you are running in a container, does restarting the container help? Might be this: https://github.com/ollama/ollama/blob/main/docs/troubleshooting.mdx#linux-docker

<!-- gh-comment-id:3804368275 --> @rick-github commented on GitHub (Jan 27, 2026): It looks like you are running in a container, does restarting the container help? Might be this: https://github.com/ollama/ollama/blob/main/docs/troubleshooting.mdx#linux-docker
GiteaMirror commented 2026-04-29 09:30:24 -05:00
Author
Owner
Copy Link

@discostur commented on GitHub (Jan 27, 2026):

@rick-github ok seems like a simple reboot fixed it ... not sure why ;) thanks anyway

<!-- gh-comment-id:3804617474 --> @discostur commented on GitHub (Jan 27, 2026): @rick-github ok seems like a simple reboot fixed it ... not sure why ;) thanks anyway
GiteaMirror commented 2026-04-29 09:30:25 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jan 27, 2026):

A reboot restarted the container. If ollama starts using the CPU instead of the GPU again, see the link.

<!-- gh-comment-id:3804654408 --> @rick-github commented on GitHub (Jan 27, 2026): A reboot restarted the container. If ollama starts using the CPU instead of the GPU again, see the link.
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#55628
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?