lllyasviel / lllyasviel/stable-diffusion-webui-forge
Patching LoRA on fp16 unet && fp16 T5 really necessary?
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 13k
- Forks
- 1.7k
- PR merge metrics
- No merged PRs in 30d
Description
Why do I still see the "Patching LoRA" message when using fp16 unet and fp16/fp8 T5XXL?
Is this necessary? Could the problem be that LoRA is saved in bf16?
fp8 unet doesn't have this problem. No patching is needed.
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
No files, tests, or entry points are named. Reproduce the message with an fp16 UNet and fp16/fp8 T5XXL, compare it with an fp8 UNet, and investigate whether the LoRA bf16 format explains the patching; done means establishing whether the behavior is necessary and documenting or correcting it.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100