Comfy-Org / Comfy-Org/ComfyUI

Windows + AMD RX 6700 XT: ComfyUI crashes on Generate (exit code 0xC0000005)

Open
#11,931 5 comments 0 reactions 0 assignees View on GitHub
Potential Bug
Dominant language
Python
Stars
133k
Forks
15.7k
Avg merge
1d 7h
Merged PRs (30d)
158

Description

### Custom Node Testing

- [ ] I have tried disabling custom nodes and the issue persists (see [how to disable custom nodes](https://docs.comfy.org/troubleshooting/custom-node-issues#step-1%3A-test-with-all-custom-nodes-disabled) if you need help)

### Expected Behavior

When clicking "Generate", sampling should start and progress steps should appear in the console.
An image should be generated successfully.

### Actual Behavior

When clicking "Generate", the backend process exits immediately before sampling starts.
No Python traceback is shown.
The UI switches to "Reconnecting...".
The process exits with code -1073741819 (0xC0000005).

### Steps to Reproduce

1. Start ComfyUI on Windows using the standalone AMD build.
2. Open the UI in a browser.
3. Load a simple SD 1.5 workflow (512x512, low steps).
4. Click "Generate".

### Debug Logs

```powershell
got prompt
model weight dtype torch.float16, manual cast: None
model_type EPS
Using split attention in VAE
Using split attention in VAE
VAE load device: cuda:0, offload device: cpu, dtype: torch.float32
Requested to load SD1ClipModel
loaded completely; 235.84 MB loaded, full load: True
CLIP/text encoder model load device: cpu, offload device: cpu, current: cpu, dtype: torch.float16
Requested to load BaseModel
loaded completely; 10700.86 MB usable, 1639.41 MB loaded, full load: True

Process exit code:
-1073741819 (0xC0000005, Access Violation)
```

### Other

Already tried:

- --force-fp16
- --use-split-cross-attention
- --lowvram
- --disable-xformers
- HSA_ENABLE_SDMA=0
- SD 1.5 base checkpoint
- Low resolution (512x512) and low steps

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.