Instructions to use huaichang/PersonaLive with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use huaichang/PersonaLive with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline from diffusers.utils import load_image, export_to_video # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("huaichang/PersonaLive", dtype=torch.bfloat16, device_map="cuda") pipe.to("cuda") prompt = "A man with short gray hair plays a red electric guitar." image = load_image( "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/guitar-man.png" ) output = pipe(image=image, prompt=prompt).frames[0] export_to_video(output, "output.mp4") - Notebooks
- Google Colab
- Kaggle
Works great on an RTX 6000 Pro
Only had to make a few small changes for sm_120, but worked after.
Suggestion:
Image cannot be changed unless app is restarted. Also, optimization options like resolution and others (in the webUI) so we can get less delay.
Really nice app, I love it.
One more thing, I spent hours trying to get TRT to work, but can't get it working.
Only had to make a few small changes for sm_120, but worked after.
Suggestion:
Image cannot be changed unless app is restarted. Also, optimization options like resolution and others (in the webUI) so we can get less delay.Really nice app, I love it.
Thank you for your feedback and support! I'm glad to hear you like the app. 🥳
Regarding your suggestions (changing the reference image without restarting and adding resolution/optimization options), I have added them to my TODO list and will work on them in future updates.
Could you please share more details about the "small changes" you made for sm_120? It would be very helpful if you could paste the changes here so I can improve compatibility for everyone. Thanks again!
One more thing, I spent hours trying to get TRT to work, but can't get it working.
What specific error are you encountering? If you can share the error message here, I might be able to offer some clues.
Would really be helpful if you can assist with setting up to run on a 5090 build as im facing same sm_120 error during installation.
I'll be honest, I don't remember how I did it. I just used Google Gemini to tell me what to do.
Haha. Well would be helpful if you can make a full zip of your installation folder and upload on a Google Drive link and share so we can all decompile and figure out the tweaks. The repo owner even wants to know how it worked