Congrats
#1
by
patrickvonplaten
- opened
Hey @roygan ,
Congrats on releasing the new model! The examples look great :-)
It might be interesting to also add Flax weights to this model to allow for super fast inference on Google Colab. It should be as simple as opening a new branch called "flax" and then pushing the flax weights that can be retrieved via:
from diffusers import FlaxStableDiffusionPipeline
pipeline, params = FlaxStableDiffusionPipeline.from_pretrained("IDEA-CCNL/Taiyi-Stable-Diffusion-1B-Chinese-v0.1", from_pt=True)
pipeline.save_pretrained("<path/to/hub/repo/with/flax/branch>", params=params)
Happy to help if you'd like :-)
Thanks for your advice,we will add the flax weights soon!
wuxiaojun
changed discussion status to
closed