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 #7016] Ollama not utilizing Radeon RX 570 GPU - Possible incompatibility? #66503

New Issue
Closed
opened 2026-05-04 07:02:52 -05:00 by GiteaMirror · 3 comments
GiteaMirror commented 2026-05-04 07:02:52 -05:00
Owner
Copy Link

Originally created by @X4n71uM on GitHub (Sep 28, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/7016

What is the issue?

Hi Ollama team and community,

I'm running Ollama (v0.3.12) on Windows 10 and I'm unable to get it to utilize my Radeon RX 570 OC 8GB GPU (With latest Adrenaline Driver). Ollama seems to be defaulting to CPU, which is significantly slower for my use case.

Here's the relevant part of the log:
"""
time=2024-09-28T02:43:51.626+02:00 level=INFO source=gpu.go:199 msg="looking for compatible GPUs"
time=2024-09-28T02:43:51.744+02:00 level=INFO source=gpu.go:347 msg="no compatible GPUs were discovered"
"""

Is there any way to force GPU usage, perhaps through a command-line flag or configuration setting? Is my RX 570 simply too old or unsupported by Ollama's current ROCm implementation?

I'd appreciate any guidance on how to enable GPU usage or confirmation if my GPU is indeed incompatible. Even if my GPU isn't ideal for large models, any performance boost would be welcome.

Thanks in advance!

OS

Windows

GPU

AMD

CPU

AMD

Ollama version

v0.3.12

Originally created by @X4n71uM on GitHub (Sep 28, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/7016 ### What is the issue? Hi Ollama team and community, I'm running Ollama (v0.3.12) on Windows 10 and I'm unable to get it to utilize my Radeon RX 570 OC 8GB GPU (With latest Adrenaline Driver). Ollama seems to be defaulting to CPU, which is significantly slower for my use case. Here's the relevant part of the log: """ time=2024-09-28T02:43:51.626+02:00 level=INFO source=gpu.go:199 msg="looking for compatible GPUs" time=2024-09-28T02:43:51.744+02:00 level=INFO source=gpu.go:347 msg="no compatible GPUs were discovered" """ Is there any way to force GPU usage, perhaps through a command-line flag or configuration setting? Is my RX 570 simply too old or unsupported by Ollama's current ROCm implementation? I'd appreciate any guidance on how to enable GPU usage or confirmation if my GPU is indeed incompatible. Even if my GPU isn't ideal for large models, any performance boost would be welcome. Thanks in advance! ### OS Windows ### GPU AMD ### CPU AMD ### Ollama version v0.3.12
GiteaMirror added the bug label 2026-05-04 07:02:53 -05:00
GiteaMirror commented 2026-05-04 07:06:46 -05:00
Author
Owner
Copy Link

@dajamox commented on GitHub (Sep 28, 2024):

ROCm isn't really supported on Polaris, you might want to look out for #5059 since that adds Vulkan, which should work.

<!-- gh-comment-id:2381008283 --> @dajamox commented on GitHub (Sep 28, 2024): ROCm isn't really supported on Polaris, you might want to look out for #5059 since that adds Vulkan, which should work.
GiteaMirror commented 2026-05-04 07:06:51 -05:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (Sep 28, 2024):

Older Radeon support is being tracked via #2453

<!-- gh-comment-id:2381014959 --> @dhiltgen commented on GitHub (Sep 28, 2024): Older Radeon support is being tracked via #2453
GiteaMirror commented 2026-05-04 07:06:54 -05:00
Author
Owner
Copy Link

@juwonpee commented on GitHub (Nov 19, 2024):

Wouldnt vulkan still be supported though?

<!-- gh-comment-id:2484792772 --> @juwonpee commented on GitHub (Nov 19, 2024): Wouldnt vulkan still be supported though?
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#66503
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?