huggingface/diffusers

Support for InfiniteTalk

Open

#12239 opened on Aug 26, 2025

View on GitHub
 (1 comment) (2 reactions) (0 assignees)Python (22,190 stars) (4,562 forks)batch import
New pipeline/modelcontributions-welcomehelp wanted

Description

Model/Pipeline/Scheduler description

https://huggingface.co/MeiGen-AI/InfiniteTalk is a wonderful audio driven video generation model and can also support infinite frame , which is based on wan2.1. The demo and user's workflow is also awesome. some examples: https://www.runninghub.cn/ai-detail/1958438624956203010

Open source status

  • The model implementation is available.
  • The model weights are available (Only relevant if addition is not a scheduler).

Provide useful links for the implementation

https://huggingface.co/MeiGen-AI/InfiniteTalk https://github.com/MeiGen-AI/InfiniteTalk

Contributor guide

Tech stack
pythonpytorch
Domain
aimachine learning
Issue type
feature
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
4
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-2 days
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
fresh
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
clear
Prerequisites
Python and PyTorch experienceFamiliarity with diffusers libraryUnderstanding of diffusion modelsAudio driven video generation concepts
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
30
Research direction
Examine the InfiniteTalk repository at https://github.com/MeiGen AI/InfiniteTalk to understand the model architecture (based on Wan2.1). Study the diffusers codebase, particularly existing video generation pipelines, to determine how to integrate the model. Create a new pipeline class that loads the InfiniteTalk model, handles audio input and video output, and adds necessary preprocessing/postprocessing. Test the implementation using the provided weights from Hugging Face. Refer to related issues or PRs in diffusers for similar integrations.