Samit

Thank you for your contribution to the MindCV repo. Before submitting this PR, please make sure: - [x] You have read the [Contributing Guidelines on pull requests](https://github.com/mindspore-lab/mindcv/blob/main/CONTRIBUTING.md) - [x] Your...

[RFC] MindCV 0.2 Roadmap - New Feature/Model Request

7

We are open to collecting valuable new models and algorithms from users! You are welcome to leave a comment to suggest a new feature/model/algorithm for MIndCV. You can also vote...

PLEASE FILL IN THE PR DESCRIPTION HERE ENSURING ALL CHECKLIST ITEMS (AT THE BOTTOM) HAVE BEEN CONSIDERED. ## Purpose Fix #235 ## Test Plan ```shell python text_to_image.py --enable_cache_dit --num_inference_steps=50 ```...

[New Model]: Add LongCat-Image

2

### The model to consider. LongCat-Image: https://huggingface.co/meituan-longcat/LongCat-Image ### The closest model vllm-omni already supports. Qwen-Image ### What's your difficulty of supporting the model you want? _No response_ ### Use case...

### Motivation. Accelerate diffusion-only model inference with cache-dit integration (DBCache + TaylorSeer, and possibly SCM) ### Proposed Change. In the first stage, we plan to support DBCache and TaylorSeer. (possibly...

[RFC]: Support Online Serving for Diffusion-only Models (QwenImage)

3

### Motivation. Currently, QwenImage only supports offline inference. We hope to support online serving with ``` vllm serve Qwen/Qwen-Image --omni --port 8000 ``` Then users can send request like ```...