MiniMax H3 in ComfyUI: Complete Video Generation Guide

Tutorial Overview

This tutorial covers everything you need to run MiniMax H3 in ComfyUI: installing the models, loading the official workflows, understanding the three generation modes (T2V, I2V, R2V), picking the right resolution, and solving common problems.

MiniMax H3 Model Overview

MiniMax H3 is the latest model in MiniMax's Hailuo video line. It generates up to 15 seconds of 24 FPS video with native 32 kHz stereo audio, in any of 11 languages, at up to 2K resolution. The model is powered by a 33.1B dense single-stream omni transformer with a Qwen3-VL-32B text encoder.

Key capabilities:

• Native stereo audio: dialogue, sound effects, and music generated in the same pass as the video, no separate audio model
• Omni-modal context: any combination of text, images, video, and audio as input
• Three task modes: text-to-video (T2V), image-to-video (I2V), and reference-to-video (R2V)
• 11 languages: Arabic, Chinese, English, French, German, Italian, Japanese, Korean, Portuguese, Russian, Spanish
• Open weights: MiniMax H3 Community License, fully local inference

The open release covers H3-Base as two checkpoints: FL2VA (text-to-video and first/last-frame conditioning) and Ref2VA (reference-based generation). The H3-Context-IR preprocessing system and H3-Regenerate-2K 2K upscaling module remain hosted APIs.

Getting Started

MiniMax H3 is supported in ComfyUI with open weights:

1. Update ComfyUI to the latest version (0.30.0 or later; native support merged in Comfy-Org/ComfyUI #15224)
2. Open Workflow -> Browse Templates -> Video, and choose any MiniMax H3 workflow
3. Follow the pop-up to download the models and run the workflow

The model files are hosted on Hugging Face in the Comfy-Org/MiniMax-H3 repository.

Model Installation

Download the following files from Comfy-Org/MiniMax-H3 and place them in their folders:

Diffusion model (choose per mode):

Text encoder (choose one):

VAEs (both required for all workflows):

Official Workflows

Six official templates are available through Comfy-Org/workflow_templates. Three run locally (open weights) and three call the MiniMax API. This guide covers the three local workflows in detail, following the official Comfy-Org documentation.

Loading the Workflows

Method 1: Template gallery (recommended)

Update ComfyUI to the latest version, then load via Workflow -> Browse Templates -> Video -> MiniMax H3.

Method 2: Download and drag

Download the JSON and drag it into the ComfyUI window.

Setting the Output Resolution

Each workflow uses a Resolution Selector node that computes width and height from three settings:

• Aspect ratio: presets such as 16:9 (Widescreen), 9:16 (Portrait), or 1:1 (Square)
• Megapixels: target total pixel count; higher values give larger frames, lower values run faster
• Multiple: rounded to the nearest multiple of this number; keep it at 32 to match H3's resolution grid

H3's native canvas is a 768px short edge, capped at 768x1344 pixels and rounded to a multiple of 32. The template ships with a fast preview size; for full quality, raise Megapixels to about 1.0 at 16:9, which yields roughly 1344x768.

1. MiniMax H3 Text-to-Video (T2V)

Generate videos from text prompts with native stereo audio.

Model downloads (T2V)

Model storage (T2V)

ComfyUI/
├── 📂 models/
│   ├── 📂 diffusion_models/
│   │   └── minimax_h3_fl2va_pruned_int8_convrot.safetensors
│   ├── 📂 text_encoders/
│   │   └── qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors
│   └── 📂 vae/
│       ├── minimax_h3_video_vae_fp16.safetensors
│       └── minimax_h3_audio_vae_fp32.safetensors

Prompting tips (T2V)

• Describe the whole scene: state the overall scene first (location, character, what is happening), then break it into timed shots
• Shots, camera, and audio: describe the shots, camera moves, and the accompanying audio (dialogue, SFX, music) in one prompt block
• Duration: the duration input snaps to the model's 17-frame-per-block (17k+5) grid at 24 FPS
• First/last frame: connect images to first_frame and/or last_frame on the MiniMaxH3ImageToVideo node to turn this workflow into first/last-frame image-to-video

2. MiniMax H3 Image-to-Video (I2V)

Generate videos from an input image, with optional first/last-frame keyframes.

Input image: - download the default input image, or use your own.

Model downloads (I2V)

Model storage (I2V)

Same layout as the T2V workflow (FL2VA diffusion model + NVFP4 text encoder + both VAEs).

Prompting tips (I2V)

• Keyframes: the first_frame and last_frame inputs are optional; the model generates the motion between them
• Prompt: describe the shots, motion, and the accompanying audio (dialogue, SFX, music) in one block
• Duration: snaps to the 17-frame-per-block (17k+5) grid at 24 FPS

3. MiniMax H3 Reference-to-Video (R2V)

Generate videos that lock in a character, style, motion, camera move, or voice from any mix of reference images, videos, and audio.

Reference images:

Character reference for the workflow, or use your own image.

Style and subject reference for the workflow, or use your own image.

Model downloads (R2V)

Model storage (R2V)

ComfyUI/
├── 📂 models/
│   ├── 📂 diffusion_models/
│   │   └── minimax_h3_ref2va_pruned_int8_convrot.safetensors
│   ├── 📂 text_encoders/
│   │   └── qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors
│   └── 📂 vae/
│       ├── minimax_h3_video_vae_fp16.safetensors
│       └── minimax_h3_audio_vae_fp32.safetensors

Prompting tips (R2V)

• Reference by tag: reference each input by tag in the exact order it was connected, for example <Picture 1>, <Video 1>, <Audio 1>
• Assign each reference a job: state which reference drives which part of the shot (identity, style, motion, camera, voice); explicit assignments work much better
• Limits: up to 9 reference images, 3 reference videos (each can carry its own soundtrack), and 3 standalone reference audio clips
• ref_image_size: match scales references down to the generation resolution for speed; max keeps up to a 2048px short edge for stronger identity fidelity at the cost of speed
• Sampler: res_multistep with a beta or normal scheduler tends to outperform simple for reference-heavy prompts
• Note: R2V uses the ref2va diffusion model, a different set of weights from the fl2va model used by T2V and I2V

API Workflows

Three additional official templates call the MiniMax API instead of running locally:

Troubleshooting

Q: ComfyUI doesn't show the MiniMax H3 nodes

Update ComfyUI to version 0.30.0 or later. Native support merged in Comfy-Org/ComfyUI #15224 on August 3, 2026. Desktop and Cloud updates follow stable releases, so some nightly-supported features may not be available yet.

Q: Out of memory / high VRAM usage

• Use the pruned INT8 diffusion model (19.5 GB) instead of bf16
• Use the NVFP4 AWQ text encoder (14.6 GB), which runs on any GPU
• Lower the resolution in the Resolution Selector node (lower Megapixels) and shorten the duration
• ComfyUI's dynamic VRAM system and block swap help on machines with ample system RAM

Q: Generation fails at 256p or very small resolutions

H3 has a minimum resolution of 384p. 256p fails completely. Use the resolution presets from the official template.

Q: Output video has no audio

Make sure both VAEs are loaded: minimax_h3_video_vae_fp16.safetensors (video) and minimax_h3_audio_vae_fp32.safetensors (audio), and that the workflow includes a VAEDecodeAudio node connected to SaveVideo.

Q: Which checkpoint should I use for image-to-video?

Use the FL2VA checkpoint for T2V and I2V (first-frame, last-frame, or both). Use the Ref2VA checkpoint when you need full reference-based generation (identity, style, voice).

Q: Where can I report bugs?

• Runtime errors: ComfyUI/issues
• UI issues: ComfyUI_frontend/issues

Related Resources

• MiniMax H3 open weights news
• MiniMax H3 model page
• Official Comfy-Org MiniMax H3 documentation
• MiniMax H3 on Hugging Face