github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #10014] Podman on Ubuntu can't find gpus #68623

New Issue
Closed
opened 2026-05-04 14:38:30 -05:00 by GiteaMirror · 6 comments
GiteaMirror commented 2026-05-04 14:38:30 -05:00
Owner
Copy Link

Originally created by @thezachdrake on GitHub (Mar 27, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/10014

What is the issue?

When I run with podman on ubuntu it cannot find my gpus.

Relevant log output

nvidia-ctk cdi list

INFO[0000] Found 3 CDI devices                          
nvidia.com/gpu=0
nvidia.com/gpu=GPU-0de6d016-e195-8065-ed7d-8527f69bc2a2
nvidia.com/gpu=all


----------------------------------------------------------------
podman run -it -e NVIDIA_VISIBLE_DEVICES=all --device nvidia.com/gpu=all ollama/ollama

Couldn't find '/root/.ollama/id_ed25519'. Generating new private key.
Your new public key is: 

ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIBDuyHDev/cny2FiG1bUtT7CroPIRxQkvpdXJfZDfHpx

2025/03/27 13:54:22 routes.go:1230: INFO server config env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:2048 OLLAMA_DEBUG:false OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://0.0.0.0:11434 OLLAMA_INTEL_GPU:false OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/root/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:0 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES: http_proxy: https_proxy: no_proxy:]"
time=2025-03-27T13:54:22.051Z level=INFO source=images.go:432 msg="total blobs: 0"
time=2025-03-27T13:54:22.051Z level=INFO source=images.go:439 msg="total unused blobs removed: 0"
time=2025-03-27T13:54:22.051Z level=INFO source=routes.go:1297 msg="Listening on [::]:11434 (version 0.6.2)"
time=2025-03-27T13:54:22.052Z level=INFO source=gpu.go:217 msg="looking for compatible GPUs"
time=2025-03-27T13:54:22.059Z level=WARN source=gpu.go:605 msg="unknown error initializing cuda driver library /usr/lib/x86_64-linux-gnu/libcuda.so.550.120: cuda driver library init failure: 999. see https://github.com/ollama/ollama/blob/main/docs/troubleshooting.md for more information"
time=2025-03-27T13:54:22.064Z level=INFO source=gpu.go:377 msg="no compatible GPUs were discovered"
time=2025-03-27T13:54:22.064Z level=INFO source=types.go:130 msg="inference compute" id=0 library=cpu variant="" compute="" driver=0.0 name="" total="31.0 GiB" available="20.8 GiB"

OS

Linux

GPU

Nvidia

CPU

Intel

Ollama version

No response

Originally created by @thezachdrake on GitHub (Mar 27, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/10014 ### What is the issue? When I run with podman on ubuntu it cannot find my gpus. ### Relevant log output ```shell nvidia-ctk cdi list INFO[0000] Found 3 CDI devices nvidia.com/gpu=0 nvidia.com/gpu=GPU-0de6d016-e195-8065-ed7d-8527f69bc2a2 nvidia.com/gpu=all ---------------------------------------------------------------- podman run -it -e NVIDIA_VISIBLE_DEVICES=all --device nvidia.com/gpu=all ollama/ollama Couldn't find '/root/.ollama/id_ed25519'. Generating new private key. Your new public key is: ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIBDuyHDev/cny2FiG1bUtT7CroPIRxQkvpdXJfZDfHpx 2025/03/27 13:54:22 routes.go:1230: INFO server config env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:2048 OLLAMA_DEBUG:false OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://0.0.0.0:11434 OLLAMA_INTEL_GPU:false OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/root/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:0 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES: http_proxy: https_proxy: no_proxy:]" time=2025-03-27T13:54:22.051Z level=INFO source=images.go:432 msg="total blobs: 0" time=2025-03-27T13:54:22.051Z level=INFO source=images.go:439 msg="total unused blobs removed: 0" time=2025-03-27T13:54:22.051Z level=INFO source=routes.go:1297 msg="Listening on [::]:11434 (version 0.6.2)" time=2025-03-27T13:54:22.052Z level=INFO source=gpu.go:217 msg="looking for compatible GPUs" time=2025-03-27T13:54:22.059Z level=WARN source=gpu.go:605 msg="unknown error initializing cuda driver library /usr/lib/x86_64-linux-gnu/libcuda.so.550.120: cuda driver library init failure: 999. see https://github.com/ollama/ollama/blob/main/docs/troubleshooting.md for more information" time=2025-03-27T13:54:22.064Z level=INFO source=gpu.go:377 msg="no compatible GPUs were discovered" time=2025-03-27T13:54:22.064Z level=INFO source=types.go:130 msg="inference compute" id=0 library=cpu variant="" compute="" driver=0.0 name="" total="31.0 GiB" available="20.8 GiB" ``` ### OS Linux ### GPU Nvidia ### CPU Intel ### Ollama version _No response_
GiteaMirror added the bugneeds more info labels 2026-05-04 14:38:31 -05:00
GiteaMirror commented 2026-05-04 14:38:34 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Mar 27, 2025):

What does the following output:

podman run --rm --device nvidia.com/gpu=all --security-opt=label=disable ubuntu nvidia-smi -L
<!-- gh-comment-id:2758394863 --> @rick-github commented on GitHub (Mar 27, 2025): What does the following output: ``` podman run --rm --device nvidia.com/gpu=all --security-opt=label=disable ubuntu nvidia-smi -L ```
GiteaMirror commented 2026-05-04 14:38:36 -05:00
Author
Owner
Copy Link

@thezachdrake commented on GitHub (Mar 27, 2025):

podman run --rm --device nvidia.com/gpu=all --security-opt=label=disable ubuntu nvidia-smi -L
Resolved "ubuntu" as an alias (/etc/containers/registries.conf.d/shortnames.conf)
Trying to pull docker.io/library/ubuntu:latest...
Getting image source signatures
Copying blob 5a7813e071bf done   | 
Copying config a04dc4851c done   | 
Writing manifest to image destination
GPU 0: NVIDIA RTX A1000 Laptop GPU (UUID: GPU-0de6d016-e195-8065-ed7d-8527f69bc2a2)
<!-- gh-comment-id:2758584223 --> @thezachdrake commented on GitHub (Mar 27, 2025): ``` podman run --rm --device nvidia.com/gpu=all --security-opt=label=disable ubuntu nvidia-smi -L Resolved "ubuntu" as an alias (/etc/containers/registries.conf.d/shortnames.conf) Trying to pull docker.io/library/ubuntu:latest... Getting image source signatures Copying blob 5a7813e071bf done | Copying config a04dc4851c done | Writing manifest to image destination GPU 0: NVIDIA RTX A1000 Laptop GPU (UUID: GPU-0de6d016-e195-8065-ed7d-8527f69bc2a2) ```
GiteaMirror commented 2026-05-04 14:38:38 -05:00
Author
Owner
Copy Link

@thezachdrake commented on GitHub (Mar 27, 2025):

When I run ollama as a service and not in a container it does run on the GPU and its fast. Its only when I run it in the container that it runs on CPU and is slow

<!-- gh-comment-id:2758586937 --> @thezachdrake commented on GitHub (Mar 27, 2025): When I run ollama as a service and not in a container it does run on the GPU and its fast. Its only when I run it in the container that it runs on CPU and is slow
GiteaMirror commented 2026-05-04 14:38:41 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Mar 27, 2025):

OK, so the device is available. What if you run nvidia-smi -L inside the ollama container?

<!-- gh-comment-id:2758624049 --> @rick-github commented on GitHub (Mar 27, 2025): OK, so the device is available. What if you run `nvidia-smi -L` inside the ollama container?
GiteaMirror commented 2026-05-04 14:38:42 -05:00
Author
Owner
Copy Link

@thezachdrake commented on GitHub (Mar 27, 2025):

podman run --rm --gpus nvidia.com/all ollama/ollama nvidia-smi -L
Error: unknown command "nvidia-smi" for "ollama"

is this what you mean?

<!-- gh-comment-id:2758769823 --> @thezachdrake commented on GitHub (Mar 27, 2025): ``` podman run --rm --gpus nvidia.com/all ollama/ollama nvidia-smi -L Error: unknown command "nvidia-smi" for "ollama" ``` is this what you mean?
GiteaMirror commented 2026-05-04 14:38:42 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Mar 27, 2025):

I don't use podman, my guess would be:

podman run --rm --gpus nvidia.com/all --entrypoint bash ollama/ollama -c 'nvidia-smi -L'
<!-- gh-comment-id:2758903374 --> @rick-github commented on GitHub (Mar 27, 2025): I don't use podman, my guess would be: ``` podman run --rm --gpus nvidia.com/all --entrypoint bash ollama/ollama -c 'nvidia-smi -L' ```
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 needs more info
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#68623
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?