github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-08 00:51:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #13188] AMD 780M GPU Acceleration Not Working with 2024.12 Driver, Works with 2025.11 (Ollama 0.13.0) #70777

New Issue
Closed
opened 2026-05-04 22:57:16 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-05-04 22:57:16 -05:00
Owner
Copy Link

Originally created by @Zhucan123 on GitHub (Nov 21, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/13188

Originally assigned to: @dhiltgen on GitHub.

What is the issue?

Hi Ollama team,

I’m encountering an issue with GPU acceleration on an AMD 780M graphics card using the official Ollama 0.13.0 release.

Issue Background:

When using the AMD 780M with driver version 2024.12, Ollama does not utilize GPU acceleration; all computation falls back to the CPU.
Upgrading the AMD driver to the latest 2025.11 version resolves the issue, and GPU acceleration works as expected.
System Info:

Ollama version: 0.13.0
GPU: AMD 780M
OS: [please specify, e.g. Windows 11, Ubuntu 22.04, etc.]
Driver version tested:
2024.12 (No GPU acceleration)
2025.11 (GPU acceleration works)
Steps to Reproduce:

Install AMD 780M graphics card and set driver version to 2024.12
Install Ollama 0.13.0
Attempt model inference; observe that GPU is not utilized
Upgrade driver to 2025.11
Retry inference; GPU acceleration works correctly
Expected Behavior:

Ollama should utilize GPU acceleration on AMD 780M regardless of using either the 2024.12 or 2025.11 drivers.

Actual Behavior:

GPU acceleration only works on 2025.11; does not work on 2024.12.

Logs & Additional Info:

[Attach logs or screenshots if available.]

Could you please help investigate why the older 2024.12 driver is not supported for GPU acceleration, and if possible, provide a fix or workaround?

Thanks!

Relevant log output

log.txt

OS

amd windows

GPU

radeon 780m

CPU

ryzen 7 8745HS w

Ollama version

0.13.0

Originally created by @Zhucan123 on GitHub (Nov 21, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/13188 Originally assigned to: @dhiltgen on GitHub. ### What is the issue? Hi Ollama team, I’m encountering an issue with GPU acceleration on an AMD 780M graphics card using the official Ollama 0.13.0 release. Issue Background: When using the AMD 780M with driver version 2024.12, Ollama does not utilize GPU acceleration; all computation falls back to the CPU. Upgrading the AMD driver to the latest 2025.11 version resolves the issue, and GPU acceleration works as expected. System Info: Ollama version: 0.13.0 GPU: AMD 780M OS: [please specify, e.g. Windows 11, Ubuntu 22.04, etc.] Driver version tested: 2024.12 (No GPU acceleration) 2025.11 (GPU acceleration works) Steps to Reproduce: Install AMD 780M graphics card and set driver version to 2024.12 Install Ollama 0.13.0 Attempt model inference; observe that GPU is not utilized Upgrade driver to 2025.11 Retry inference; GPU acceleration works correctly Expected Behavior: Ollama should utilize GPU acceleration on AMD 780M regardless of using either the 2024.12 or 2025.11 drivers. Actual Behavior: GPU acceleration only works on 2025.11; does not work on 2024.12. Logs & Additional Info: [Attach logs or screenshots if available.] Could you please help investigate why the older 2024.12 driver is not supported for GPU acceleration, and if possible, provide a fix or workaround? Thanks! ### Relevant log output [log.txt](https://github.com/user-attachments/files/23670100/log.txt) ```shell ``` ### OS amd windows ### GPU radeon 780m ### CPU ryzen 7 8745HS w ### Ollama version 0.13.0
GiteaMirror added the amdbug labels 2026-05-04 22:57:16 -05:00
GiteaMirror commented 2026-05-04 22:57:19 -05:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (Nov 21, 2025):

Your iGPU is a gfx1103 which is not supported by ROCm. From the logs, it looks like Vulkan is crashing during initialization. It's unclear exactly where the bug is from the logs, but each AMD Driver comes with a matching Vulkan library, so given upgrading the driver fixed the bug, it seems plausible the bug was in somewhere in the AMD stack and was fixed by AMD.

<!-- gh-comment-id:3564707746 --> @dhiltgen commented on GitHub (Nov 21, 2025): Your iGPU is a gfx1103 which is not supported by ROCm. From the logs, it looks like Vulkan is crashing during initialization. It's unclear exactly where the bug is from the logs, but each AMD Driver comes with a matching Vulkan library, so given upgrading the driver fixed the bug, it seems plausible the bug was in somewhere in the AMD stack and was fixed by AMD.
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 amd 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#70777
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?