Nunchaku Kontext official workflow 10-second sampling
FLUX.1 Kontext is now supported
https://modelscope.cn/models/Lmxyy1999/nunchaku flux.1 kontext dev for download, example workflow https://github.com/mit han lab/ComfyUI nunchaku/blob/main/example_workflows/nunchaku flux.1 kontext dev.json. Also compatible with Turbo LoRA: https://github.com/mit han lab/ComfyUI nunchaku/blob/main/example_workflows/nunchaku flux.1 kontext dev turbo_lora.json

Nunchaku summer development plan has been updated https://github.com/mit han lab/nunchaku/pull/418, feel free to comment!
If you are interested in model quantization, machine learning systems, and want to participate in Nunchaku development, feel free to contact the group owner: muyangli@mit.edu

Model links:
HuggingFace: https://huggingface.co/collections/mit han lab/nunchaku 6837e7498f680552f7bbb5ad
ModelScope: https://modelscope.cn/collections/Nunchaku 519fed7f9de94e

For those who encounter issues with Nunchaku installation and usage, check out this video: 【Nunchaku Windows Installation and Usage Tutorial】 https://www.bilibili.com/video/BV1BTocYjEk5/?share_source=copy_web&vd_source=8926212fef622f25cc95380515ac74ee

🔧Common issues and solutions can be found here:
👉https://github.com/mit han lab/nunchaku/blob/main/docs/faq.md