nkxx188/ComfyUI-MiniMaxH3-Easy

The easiest way to use MiniMax H3. One compact workflow for T2V, I2V, first/last-frame, and reference video generation, with a unified multi-media input, powerful @ references, and inline dialogue blocks. Less wiring. More creative control.

What it solves

This project provides a simplified, high-level interface for using the MiniMax H3 video generation model within ComfyUI. It replaces complex, multi-node setups with a compact "Easy" node that manages prompts, media inputs, and model configurations, while remaining compatible with the rest of the ComfyUI ecosystem for sampling and post-processing.

How it works

The system centers around a main "Easy" node that handles the core logic of text-to-video, image-to-video, and reference-based video generation.

  • Media Handling: It uses a unique multi-link Media port that allows images, videos, and audio to be connected to a single input, which the node then numbers and tracks independently.
  • Prompting: It features a structured editor supporting @ references to connected media and # dialogue blocks for speech/lyrics, which are converted into the specific tags required by MiniMax H3.
  • Prompt Optimization: An optional built-in tool can rewrite prompts using external APIs (OpenAI, Gemini) based on mode-aware guides to improve generation quality.
  • Model Management: A bundled loader handles the necessary transformers, text encoders, and VAEs, though a "Model Bridge" is provided for users who prefer using their own loaders (including GGUF variants).
  • Two-Pass Workflow: It includes a specialized second-pass conditioning node to allow for high-resolution refinement without the keyframe mismatch issues typically found when changing resolutions between passes.

Who it’s for

Users of ComfyUI who want to access MiniMax H3's video generation capabilities without manually configuring dozens of nodes, and creators who need precise control over reference media and dialogue in their AI videos.

Highlights

  • Unified Media Port: Connect multiple images, videos, and audio clips to one port with distinct wire colors and quick-create menus.
  • @ Reference Editor: Easily tag connected media within the prompt using a picker.
  • Dialogue Blocks: Specialized syntax for creating dialogue and lyrics that are preserved during generation.
  • API-Driven Prompt Optimizer: Integrated support for OpenAI and Gemini to rewrite prompts using specific H3 scene guides.
  • Pass 2 Refinement: Dedicated tools for resolution-changing second passes to maintain composition and timing.
  • Flexible Model Loading: Supports both bundled loading and custom model bridges for GGUF and safetensors files.

Related

  • Project
    huangserva/ComfyUI_MiniMaxH3_Director

    A collection of ComfyUI workflows for MiniMax H3 Director, enabling text-to-video, image-to-video, and advanced video-to-video editing with reference materials.

  • Dispatch
    MiniMax H3 Support in ComfyUI

    ComfyUI introduces day-zero support for MiniMax H3, an open-weights omni-modal video model capable of generating 2K video with native stereo audio on consumer hardware.

  • Project
    AIMixer/ComfyUI_MiniMaxH3_Director

    A ComfyUI plugin that provides a director's console for MiniMax H3, enabling multi-segment video planning, synchronized audio generation, and advanced refinement tools for long-form AI video.

  • Project
    Carasibana/ComfyUI-H3-FaceRefine

    A ComfyUI custom node set that improves the quality of small faces in MiniMax H3 videos by detecting, cropping, regenerating, and stitching them back into the original clip.

  • Project
    vrgamegirl19/comfyui-vrgamedevgirl

    A collection of ComfyUI custom nodes featuring an AI Video Builder for scene-by-scene production, video enhancement, and LoRA training for AI video and music video creation.