Friendly InfiniteTalk Pure

Welcome to my 💫🎶 Friendly InfiniteTalk Pure!

✨ Less mess, more magic

You've probably already heard about the high quality lipsync model InfiniteTalk that works with Kijai WanVideoWrapper. But this is the special one! InfiniteTalk Pure based on new native ComfyUI nodes with better optimization. I've made the workflow with friendly UI and additional perfomance optimizations using MultiGPU and new Kijai Chunk FeedForward node.

🚀 Don't forget to update ComfyUI and nodes in the workflow for better performance!

💻 System requirements:

• Minimum system requirements for 540p:

RTX 3000-s, 8GB+ VRAM, 45GB+ RAM, 8-core processor, SSD, latest ComfyUI

🚀 Low VRAM optional optimization:

• For systems with low VRAM use --reserve-vram ComfyUI parameter in run_nvidia_gpu.bat: --reserve-vram 4 (or other number in GB).

📌 Detailed tips and links in the workflow

✨ Workflow features:

• Extremely user-friendly interface

• Maximum performance and optimization from 8GB of VRAM

• Convenient one-click mode switching

• Generation time setting by audio crop

• Detailed tips and links to all necessary models

• Manual random seed for complete control over generations

• Interpolation module for the smooth video

🤗🙏🏼 Thanks to MeiGen-AI and kijai

Original repo — GitHub