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 #1888] nvmlInit_v2 unable to detect Nvidia GPU in WSL #47594

New Issue
Closed
opened 2026-04-28 04:26:17 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-04-28 04:26:17 -05:00
Owner
Copy Link

Originally created by @taweili on GitHub (Jan 10, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/1888

Originally assigned to: @dhiltgen on GitHub.

Ollama has switched to using NVML to detect the Nvidia environment. However, this method failed on WSL. Here is a short C code to validate the behavior.

The nvmlReturn_t returns 9 NVML_ERROR_DRIVER_NOT_LOADED = 9. This may make sense according to the implementation of Nvidia in WSL as it use the driver from Windows host. I can't find any document on this, one way or another.

This issue prevents Ollama v0.1.18 and 0.1.19 from using Nvidia hardware in WSL.

#include <stdio.h>
#include "gpu_info_cuda.h"

cuda_init_resp_t resp;
mem_info_t mem_info;

void main(void)
{
    nvmlReturn_t ret;
    cuda_init(&resp);
    ret = resp.ch.initFn();
    printf("%d\n", ret);
}
Originally created by @taweili on GitHub (Jan 10, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/1888 Originally assigned to: @dhiltgen on GitHub. Ollama has switched to using [NVML](https://developer.nvidia.com/nvidia-management-library-nvml) to detect the Nvidia environment. However, this method failed on WSL. Here is a short C code to validate the behavior. The `nvmlReturn_t` returns 9 [NVML_ERROR_DRIVER_NOT_LOADED = 9](https://docs.nvidia.com/deploy/nvml-api/group__nvmlDeviceEnumvs.html#group__nvmlDeviceEnumvs_1g06fa9b5de08c6cc716fbf565e93dd3d0). This may make sense according to the implementation of Nvidia in WSL as it use the driver from Windows host. I can't find any document on this, one way or another. This issue prevents Ollama v0.1.18 and 0.1.19 from using Nvidia hardware in WSL. ```c #include <stdio.h> #include "gpu_info_cuda.h" cuda_init_resp_t resp; mem_info_t mem_info; void main(void) { nvmlReturn_t ret; cuda_init(&resp); ret = resp.ch.initFn(); printf("%d\n", ret); } ```
GiteaMirror added the bug label 2026-04-28 04:26:17 -05:00
GiteaMirror commented 2026-04-28 04:26:21 -05:00
Author
Owner
Copy Link

@taweili commented on GitHub (Jan 10, 2024):

This is fixed with the pull request #1897 to set the collect dynamic library is used in WSL.

<!-- gh-comment-id:1884964637 --> @taweili commented on GitHub (Jan 10, 2024): This is fixed with the pull request #1897 to set the collect dynamic library is used in WSL.
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#47594
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?