huggingface / huggingface/diffusers
Mage-Flow
- Dominant language
- Python
- Stars
- 34.5k
- Forks
- 7.3k
- Avg merge
- 3d 3h
- Merged PRs (30d)
- 91
Description
### Model/Pipeline/Scheduler description
Series of image generation models
Note: official implementation is hardcoded to CUDA and flash-attn 2 and may not be easy to port
### Open source status
- [x] The model implementation is available.
- [x] The model weights are available (Only relevant if addition is not a scheduler).
### Provide useful links for the implementation
https://github.com/microsoft/Mage/tree/main/mage_flow
https://huggingface.co/microsoft/Mage-Flow
https://huggingface.co/microsoft/Mage-Flow-Base
https://huggingface.co/microsoft/Mage-Flow-Turbo
https://huggingface.co/microsoft/Mage-Flow-Edit
https://huggingface.co/microsoft/Mage-Flow-Edit-Base
https://huggingface.co/microsoft/Mage-Flow-Edit-Turbo
Contributor guide
Research direction
Start by reading the upstream implementation in microsoft/Mage’s mage_flow directory and compare it with the linked Hugging Face Mage-Flow checkpoints. Determine the scope needed to add Mage-Flow support to diffusers, including the listed model variants, and verify that the result works despite the upstream CUDA and flash-attn 2 assumptions.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 45/100