flux fill onereward int8-convrot generated Artifact during inpaint
- Dominant language
- Python
- Stars
- 133k
- Forks
- 15.7k
- Avg merge
- 1d 7h
- Merged PRs (30d)
- 158
Description
### Custom Node Testing
- [x] 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
int8-convrot better or as fp8 model
### Actual Behavior
flux fill onereward int8-convrot generated Artifact during inpaint
### Steps to Reproduce
1. original model: https://huggingface.co/Comfy-Org/OneReward_repackaged/blob/main/split_files/diffusion_models/flux.1-fill-dev-OneReward-transformer_bf16.safetensors
2. convert script: https://github.com/Comfy-Org/comfy-model-tools/blob/main/quant_int8_convrot.py
3. no parameter used during convert
4. change to int8-convrot model in dit loader(official), changed dtype to default
### Debug Logs
```powershell
...
```
### Other
not sure where the the problem is, script/comfyui/parameter?
Contributor guide
Research direction
Start with the linked comfy-model-tools/quant_int8_convrot.py conversion script and reproduce the reported flux fill inpaint case using the original OneReward bf16 model. Then inspect the official ComfyUI dit loader with the converted int8-convrot model and default dtype; done means identifying whether the artifact comes from conversion, loading, or dtype handling.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 38/100