github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-07-16 05:05:37 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #15968] Regression in 0.23.0 - GPU Discovery fails on Arch Linux (RTX 4090), works in 0.22.1 #87858

New Issue
Closed
opened 2026-05-10 06:27:28 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-05-10 06:27:28 -05:00
Owner
Copy Link

Originally created by @moraisaugusto on GitHub (May 4, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/15968

What is the issue?

After updating to version 0.23.0 on Arch Linux, Ollama fails to discover my NVIDIA RTX 4090, reporting total_vram="0 B" and falling back to the CPU runner. Downgrading back to 0.22.1 resolves the issue immediately with the exact same environment/drivers.

How to reproduce

  • update ollama-cuda to 0.23.0 - pacman -S ollama-cuda
  • run CUDA_VISIBLE_DEVICES=0 ollama serve
  • Check logs showing library=cpu

Additional context

I have verified that my user is in the video and render groups. nvidia-smi correctly identifies the card and UUID.

 $ nvidia-smi --version
NVIDIA-SMI version  : 595.71.05
NVML version        : 595.71
DRIVER version      : 595.71.05
CUDA Version        : 13.2

Relevant log output

CUDA_VISIBLE_DEVICES=0 ollama serve

time=2026-05-05T00:54:52.974+02:00 level=INFO source=routes.go:1782 msg="server config" env="map[CUDA_VISIBLE_DEVICES:0 GGML_VK_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:0 OLLAMA_DEBUG:INFO OLLAMA_DEBUG_LOG_REQUESTS:false OLLAMA_EDITOR: 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:/home/augusto/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NO_CLOUD: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=2026-05-05T00:54:52.974+02:00 level=INFO source=routes.go:1784 msg="Ollama cloud disabled: false"

time=2026-05-05T00:54:52.976+02:00 level=INFO source=images.go:517 msg="total blobs: 52"

time=2026-05-05T00:54:52.976+02:00 level=INFO source=images.go:524 msg="total unused blobs removed: 0"

time=2026-05-05T00:54:52.976+02:00 level=INFO source=routes.go:1847 msg="Listening on 127.0.0.1:11434 (version 0.23.0)"

time=2026-05-05T00:54:52.977+02:00 level=INFO source=runner.go:67 msg="discovering available GPUs..."

time=2026-05-05T00:54:52.977+02:00 level=WARN source=runner.go:536 msg="user overrode visible devices" CUDA_VISIBLE_DEVICES=0

time=2026-05-05T00:54:52.977+02:00 level=WARN source=runner.go:540 msg="if GPUs are not correctly discovered, unset and try again"

time=2026-05-05T00:54:52.979+02:00 level=INFO source=server.go:433 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --port 41567"

time=2026-05-05T00:54:52.997+02:00 level=INFO source=types.go:60 msg="inference compute" id=cpu library=cpu compute="" name=cpu description=cpu libdirs=ollama driver="" pci_id="" type="" total="125.5 GiB" available="105.9 GiB"

time=2026-05-05T00:54:52.997+02:00 level=INFO source=routes.go:1897 msg="vram-based default context" total_vram="0 B" default_num_ctx=4096

time=2026-05-05T00:54:53.512+02:00 level=INFO source=model_recommendations.go:179 msg="model recommendations cache sleep scheduled" wait=3h20m19.467589984s consecutive_failures=0

OS

Linux

GPU

Nvidia

CPU

Intel

Ollama version

0.23.0

Originally created by @moraisaugusto on GitHub (May 4, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/15968 ### What is the issue? After updating to version `0.23.0` on Arch Linux, Ollama fails to discover my NVIDIA RTX 4090, reporting `total_vram="0 B"` and falling back to the CPU runner. Downgrading back to `0.22.1` resolves the issue immediately with the exact same environment/drivers. ## How to reproduce * update ollama-cuda to `0.23.0` - `pacman -S ollama-cuda` * run `CUDA_VISIBLE_DEVICES=0 ollama serve` * Check logs showing `library=cpu` ## Additional context I have verified that my user is in the `video` and `render` groups. `nvidia-smi` correctly identifies the card and UUID. ``` $ nvidia-smi --version NVIDIA-SMI version : 595.71.05 NVML version : 595.71 DRIVER version : 595.71.05 CUDA Version : 13.2 ``` ### Relevant log output ```shell CUDA_VISIBLE_DEVICES=0 ollama serve time=2026-05-05T00:54:52.974+02:00 level=INFO source=routes.go:1782 msg="server config" env="map[CUDA_VISIBLE_DEVICES:0 GGML_VK_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:0 OLLAMA_DEBUG:INFO OLLAMA_DEBUG_LOG_REQUESTS:false OLLAMA_EDITOR: 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:/home/augusto/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NO_CLOUD: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=2026-05-05T00:54:52.974+02:00 level=INFO source=routes.go:1784 msg="Ollama cloud disabled: false" time=2026-05-05T00:54:52.976+02:00 level=INFO source=images.go:517 msg="total blobs: 52" time=2026-05-05T00:54:52.976+02:00 level=INFO source=images.go:524 msg="total unused blobs removed: 0" time=2026-05-05T00:54:52.976+02:00 level=INFO source=routes.go:1847 msg="Listening on 127.0.0.1:11434 (version 0.23.0)" time=2026-05-05T00:54:52.977+02:00 level=INFO source=runner.go:67 msg="discovering available GPUs..." time=2026-05-05T00:54:52.977+02:00 level=WARN source=runner.go:536 msg="user overrode visible devices" CUDA_VISIBLE_DEVICES=0 time=2026-05-05T00:54:52.977+02:00 level=WARN source=runner.go:540 msg="if GPUs are not correctly discovered, unset and try again" time=2026-05-05T00:54:52.979+02:00 level=INFO source=server.go:433 msg="starting runner" cmd="/usr/bin/ollama runner --ollama-engine --port 41567" time=2026-05-05T00:54:52.997+02:00 level=INFO source=types.go:60 msg="inference compute" id=cpu library=cpu compute="" name=cpu description=cpu libdirs=ollama driver="" pci_id="" type="" total="125.5 GiB" available="105.9 GiB" time=2026-05-05T00:54:52.997+02:00 level=INFO source=routes.go:1897 msg="vram-based default context" total_vram="0 B" default_num_ctx=4096 time=2026-05-05T00:54:53.512+02:00 level=INFO source=model_recommendations.go:179 msg="model recommendations cache sleep scheduled" wait=3h20m19.467589984s consecutive_failures=0 ``` ### OS Linux ### GPU Nvidia ### CPU Intel ### Ollama version 0.23.0
GiteaMirror added the bug label 2026-05-10 06:27:28 -05:00
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#87858
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?