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 #11542] GPU not utilized on AMD Radeon RX 9070 XT (silent fallback to CPU) #69675

New Issue
Closed
opened 2026-05-04 18:48:42 -05:00 by GiteaMirror · 3 comments
GiteaMirror commented 2026-05-04 18:48:42 -05:00
Owner
Copy Link

Originally created by @m4jst3r0 on GitHub (Jul 26, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/11542

What is the issue?

### System Information
- **OS:** Windows 11
- **GPU:** AMD Radeon RX 9070 XT
- **Ollama Version:** [Wpisz wersję, którą masz zainstalowaną]
- **GPU Driver Version:** [Wpisz wersję sterownika z AMD Adrenalin]

### Describe the bug
Ollama is not utilizing my AMD Radeon RX 9070 XT for inference. Instead, it silently falls back to using the CPU, resulting in 100% CPU and RAM usage, while the GPU remains idle. This was verified by monitoring the "Compute_0" graph in the Windows Task Manager.

### Steps to reproduce
1. Run any model (e.g., `ollama run llama3 "hello"` or a more complex task with `mixtral`).
2. Observe system resource usage in Task Manager.
3. The GPU's "Compute" engine shows 0% activity, while the CPU is maxed out.

### What I've tried
- A complete driver reinstall using DDU in Safe Mode.
- Setting the `OLLAMA_LLM_LIBRARY` environment variable to `directml.dll`.
- Testing with multiple models of various sizes.

The issue persists after all troubleshooting steps, suggesting a compatibility problem between Ollama's DirectML implementation and this new GPU architecture.

Relevant log output

OS

No response

GPU

No response

CPU

No response

Ollama version

No response

Originally created by @m4jst3r0 on GitHub (Jul 26, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/11542 ### What is the issue? ### System Information - **OS:** Windows 11 - **GPU:** AMD Radeon RX 9070 XT - **Ollama Version:** [Wpisz wersję, którą masz zainstalowaną] - **GPU Driver Version:** [Wpisz wersję sterownika z AMD Adrenalin] ### Describe the bug Ollama is not utilizing my AMD Radeon RX 9070 XT for inference. Instead, it silently falls back to using the CPU, resulting in 100% CPU and RAM usage, while the GPU remains idle. This was verified by monitoring the "Compute_0" graph in the Windows Task Manager. ### Steps to reproduce 1. Run any model (e.g., `ollama run llama3 "hello"` or a more complex task with `mixtral`). 2. Observe system resource usage in Task Manager. 3. The GPU's "Compute" engine shows 0% activity, while the CPU is maxed out. ### What I've tried - A complete driver reinstall using DDU in Safe Mode. - Setting the `OLLAMA_LLM_LIBRARY` environment variable to `directml.dll`. - Testing with multiple models of various sizes. The issue persists after all troubleshooting steps, suggesting a compatibility problem between Ollama's DirectML implementation and this new GPU architecture. ### Relevant log output ```shell ``` ### OS _No response_ ### GPU _No response_ ### CPU _No response_ ### Ollama version _No response_
GiteaMirror added the bug label 2026-05-04 18:48:42 -05:00
GiteaMirror commented 2026-05-04 18:48:44 -05:00
Author
Owner
Copy Link

@Imptovskii commented on GitHub (Jul 26, 2025):

Same problem

<!-- gh-comment-id:3123435213 --> @Imptovskii commented on GitHub (Jul 26, 2025): Same problem
GiteaMirror commented 2026-05-04 18:48:45 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jul 26, 2025):

#10430

<!-- gh-comment-id:3123494945 --> @rick-github commented on GitHub (Jul 26, 2025): #10430
GiteaMirror commented 2026-05-04 18:48:48 -05:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (Jul 31, 2025):

Closing as a dup of #10430 - we're still waiting on AMD to release ROCm 6.4 for windows which will support these GPUs.

<!-- gh-comment-id:3140458644 --> @dhiltgen commented on GitHub (Jul 31, 2025): Closing as a dup of #10430 - we're still waiting on AMD to release ROCm 6.4 for windows which will support these GPUs.
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#69675
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?