AIMixer/ComfyUI_MiniMaxH3_Director

Multi-segment MiniMax H3 Director for official ComfyUI MiniMax-H3

What it solves

This project provides a comprehensive "director's console" for the MiniMax H3 video generation model within ComfyUI. It simplifies the creation of long-form videos by integrating multi-segment planning, conditional encoding, sampling, and exporting into a single node, removing the need to manually chain numerous individual nodes for complex video sequences.

How it works

The plugin extends the official MiniMax H3 implementation by adding a high-level management layer. It allows users to define a sequence of video segments using a visual timeline. Depending on the chosen task type (Text-to-Video, Image-to-Video, First-and-Last-Frame-to-Video, Reference-to-Video, or Video-to-Video), the node handles the underlying sampling and decoding process.

Key technical mechanisms include:

  • Inter-segment Guidance: It can pin the end motion and audio of a previous segment to the start of the next to ensure smooth transitions.
  • Refine/Upscale Pipeline: A separate Refine node can be attached to perform second-pass sampling for higher quality or to upscale the video using latent or pixel-based methods.
  • Reference Groups: Users can define global reference images, audio, and prompts that are shared across multiple segments, while still allowing per-segment overrides.

Who it’s for

It is designed for ComfyUI users and AI video creators who want more granular control over long-form AI video generation, specifically those using the MiniMax H3 model to create structured narratives or complex scenes with consistent characters and audio.

Highlights

  • Integrated Timeline: Visual timeline for cutting, splitting, and managing multiple video segments with thumbnail previews.
  • Versatile Task Modes: Supports t2v, i2v, fl2v (first/last frame), r2v (reference subject), v2v (video-to-video), and rv2v (reference video-to-video).
  • Native Stereo Audio: Generates synchronized stereo audio alongside the video frames.
  • Advanced Refinement: Supports multi-pass refinement and upscaling (including NVIDIA RTX VSR and H3 latent upscaling).
  • External Group Integration: Allows third-party nodes to feed processed images or prompts into the director console via specialized group nodes.

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.

  • Project
    nkxx188/ComfyUI-MiniMaxH3-Easy

    A compact ComfyUI interface for MiniMax H3 that simplifies text-to-video, image-to-video, and reference-based video generation with a unified media input system.

  • 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
    WhatDreamsCost/WhatDreamsCost-ComfyUI

    WhatDreamsCost‑ComfyUI is a collection of custom ComfyUI nodes that add video‑ and audio‑editing capabilities (timeline editor, load/trim nodes, IC‑LoRA support, audio in‑painting, etc.) for AI‑generated media workflows.

  • 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.