LeonQ8/ComfyUI-ALLinONE-MinimaxH3

What it solves

It eliminates the need to build complex node graphs and manually connect wires when using the MiniMax H3 video pipeline in ComfyUI. Instead of hunting through multiple custom node packs to find the right workflow, users can access all H3 capabilities through a single, unified node.

How it works

The project wraps the MiniMax H3 pipeline into a single node interface with multiple operational modes. Users select a mode, provide prompts or reference materials, and trigger generation. It integrates various specialized workflows—such as text-to-video, image-to-video, and masking—into one place, and includes built-in tools for live preview, history tracking, and output management.

Who it’s for

Content creators and ComfyUI users who want to utilize the MiniMax H3 video generation models without the steep learning curve or manual effort of assembling complex node-based workflows.

Highlights

• Comprehensive Modes: Supports Text-to-Video (T2V), Image-to-Video (I2V), Reference-to-Video (R2V), and Image generation.
• Advanced Video Control: Includes features for lip-sync (Audio Drive), seamless video extension, keyframe pinning, and multi-clip chaining.
• Video Editing Tools: Features a "Compare & Stitch" tool to play clips side-by-side or export them as a single combined video.
• Specialized Utilities: Includes a character sheet generator for creating turnaround sheets from reference images and a masking tool for region-specific replacement.
• Integrated Library: A searchable history and library for managing outputs with an RTX upscale hook.