[Bugfix] Fix generation artifacts of Qwen-Image-Edit-2511 and update pipeline DiT param parsing #776
Conversation
Signed-off-by: samithuang <[email protected]>
Signed-off-by: samithuang <[email protected]>
Signed-off-by: samithuang <[email protected]>
Signed-off-by: samithuang <[email protected]>
Signed-off-by: samithuang <[email protected]>
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 8ac473c1eb
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
Signed-off-by: samithuang <[email protected]>
Signed-off-by: samithuang <[email protected]>
Signed-off-by: samithuang <[email protected]>
|
@SamitHuang self.img_norm1 = AdaLayerNorm(dim, elementwise_affine=False, eps=eps) img_modulated, img_gate1 = self.img_norm1(hidden_states, img_mod1, modulate_index) |
gcanlin
left a comment
There was a problem hiding this comment.
Teacache should also change the corresponding parameter.
Signed-off-by: samithuang <[email protected]>
@yuanheng-zhao Currently qwen-image-edit with tea-cache still haves slight artifacts. I tried to add module_index in tea_cache as well, but get the following errror. Can you PTAL? |
perhaps it's related to torch compile |
…pipeline DiT param parsing (vllm-project#776) Signed-off-by: samithuang <[email protected]>
…pipeline DiT param parsing (vllm-project#776) Signed-off-by: Samit <[email protected]>
Hey @SamitHuang , may I get the command you tested and got the torch compile exception? I added |
…pipeline DiT param parsing (vllm-project#776) Signed-off-by: samithuang <[email protected]> Signed-off-by: Chen Yang <[email protected]>
…pipeline DiT param parsing (vllm-project#776) Signed-off-by: samithuang <[email protected]>
PLEASE FILL IN THE PR DESCRIPTION HERE ENSURING ALL CHECKLIST ITEMS (AT THE BOTTOM) HAVE BEEN CONSIDERED.
Purpose
Fixed #675
Test Plan
Test Result
Before:
After:

Essential Elements of an Effective PR Description Checklist
supported_models.mdandexamplesfor a new model.BEFORE SUBMITTING, PLEASE READ https://github.com/vllm-project/vllm-omni/blob/main/CONTRIBUTING.md (anything written below this line will be removed by GitHub Actions)