Feedback
AI Ad Video Example
Loading...
comfyui minimax h3
ComfyUI MiniMax H3 renders prompt-driven clips with built-in stereo audio at up to 2K — open weights, running on your own GPU.
All Tools
Discover our comprehensive AI-powered animation toolkit
MiniMax H3
MiniMax H3 AI Video Generator
Seedance 2.5
The Future of AI Video Is Here.
Seedance 2.0
The Future of AI Video Is Here.
Veo3.1
Create Stunning Videos with Veo3.1
Kling 3.0
Next-Gen AI Video Generator
MiniMax H3 video generator
MiniMax H3 AI Video Generator
Seedance 2.0
The Future of AI Video Is Here.
Kling Motion Control
Turn reference images into amazing motion videos in minutes
What Sets the comfyui minimax h3 Pipeline Apart
Built on MiniMax's omni-modal engine, the comfyui minimax h3 workflow brings open-weight generation straight into ComfyUI. Text, stills, footage, and sound are read together in one shared context, so each render arrives with its own stereo track — dialogue, effects, and score produced in the same forward pass. Clips reach 2K at 24fps for roughly fifteen seconds, and every parameter stays exposed at the node level.
- Sound Built Into Every RenderSpeech, effects, and score are produced alongside the picture and delivered in a single MP4, so the comfyui minimax h3 pipeline never needs a separate audio pass.
- Runs Fully on Your HardwareThe comfyui minimax h3 model executes locally, leaving resolution, length, and diffusion settings open to tweak with no per-call API ceiling.
- Mix Any Reference TypeFeed text, pictures, footage, and audio into one run to pin down a face, a look, a motion, a camera path, or a voice through the comfyui minimax h3 nodes.
Setting Up the comfyui minimax h3 Workflow in Three Steps
Three quick steps take you from a fresh install to open-weight clips with built-in audio via the comfyui minimax h3 workflow.
Capabilities of the comfyui minimax h3 Workflow
From three ready-made templates and omni-modal open-weight inference to built-in stereo sound, reference locking, and optional Sage Attention speedups, the comfyui minimax h3 setup covers a full local production pipeline.
Three Ready-Made Templates
Text-to-video, image-to-video, and reference-to-video examples ship inside the comfyui minimax h3 template library, each one covering a single generation mode straight away.
One Shared Context for Every Modality
Text, pictures, footage, and sound are interpreted together by the comfyui minimax h3 model, letting every reference type feed into the same generation.
Lock Anything From a Reference
Hold a face, a style, a motion, a camera move, or a voice steady using source material — up to 9 images, 3 videos, and 3 audio clips through the comfyui minimax h3 R2V node.
Crisp On-Screen Text and Logos
Lettering and brand marks come out clean with the comfyui minimax h3 model, whose instruction following lets you describe reference relationships in plain language.
Faster Renders with Sage Attention
Add the Patch Sage Attention KJ node to the comfyui minimax h3 workflow to roughly double throughput while barely touching output quality.
A Grid for Resolution and Duration
The comfyui minimax h3 Resolution Selector derives width and height from aspect ratio and megapixels, snapped to the model's 32-multiple grid and 17-frame blocks at 24fps.
comfyui minimax h3 — Common Questions
Answers to the questions people ask most about running the MiniMax H3 model inside ComfyUI.
What exactly is the comfyui minimax h3 workflow?
It is ComfyUI's built-in port of MiniMax H3, an omni-modal generation model MiniMax published as open weights. One forward pass turns text, pictures, footage, or audio references into a video that already carries its own stereo track.
What resolution and length can it produce?
Up to 2K at 24fps for roughly fifteen seconds. The native canvas keeps a 768px short edge, tops out at 768x1344 pixels, and rounds dimensions to a multiple of 32.
Which generation modes ship with it?
Three of them: text-to-video (T2V), image-to-video (I2V) with optional first/last-frame control, and reference-to-video (R2V), which pins down a character, style, motion, camera, or voice.
Does it produce sound as well?
Yes. The comfyui minimax h3 model builds stereo audio — speech, effects, and music — alongside the picture in one pass, muxed together into a single MP4 file.
How do I get started?
Bring ComfyUI up to 0.30.0 or newer, open Template Library > Video, pick a comfyui minimax h3 workflow, and follow the pop-up to pull models from the Hugging Face Comfy-Org/MiniMax-H3 repository.
Can generation be sped up?
Yes — install SageAttention and the KJNodes pack, then insert a Patch Sage Attention KJ node between the UNETLoader and BasicGuider in the comfyui minimax h3 workflow to roughly double throughput.
Put the comfyui minimax h3 Workflow to Work
Fire up MiniMax H3 on your own machine — open weights, stereo sound baked in, and every parameter within reach. Text, image, and reference video modes are ready whenever you are.