High-resolution text, frame-guided and multimodal reference video generation. Render at 768p or 2K, guide the opening or closing frame, and mix image and video references in one shot.
Try It
Set your parameters and start generating AI videos with the workflows and inputs currently supported on this page, including text, images, video clips, or audio where supported.
Describe the video you want to generate...
Visual media on this page was generated with MiniMax H3 in Vofy Studio and illustrates the creative workflows described. It is not a benchmark or a guarantee of output.
What's New
MiniMax H3 targets resolution and reference fidelity rather than raw clip length. The capabilities below reflect what Vofy exposes in Studio today.
MiniMax H3 2K product shot
H3 renders up to 2560 by 1440, which is the highest native resolution in the Vofy video lineup. You can also anchor the opening or closing frame, so a shot begins or lands exactly where you need it instead of approximating the composition from text alone.
MiniMax H3 cinematic portrait
Mix still images and video clips as references to hold a subject, wardrobe, location or motion pattern steady across a shot. Reference images are accepted up to 30 MB in JPG, PNG, WebP, HEIC or HEIF, with a long edge up to 5760 pixels and an aspect ratio between 0.4 and 2.5.
MiniMax H3 natural landscape
H3 covers the short-form range that social and advertising work actually uses. Pick 768p for fast iteration and 2K for delivery, and set any duration between 4 and 15 seconds. Pricing scales with both, so a 4-second 768p draft costs a fraction of a 15-second 2K render.
Video generated with MiniMax H3 in Vofy Studio, from 2K product work through to cinematic and documentary footage.
2K Product Cinematography
Getting Started
Start from text alone, add a first or last frame when the composition has to be exact, or attach references when a subject, product or location must stay consistent.
Describe action, camera movement and lighting. Add reference images or clips for anything that has to remain recognisable across the shot.
Choose 768p or 2K and any duration between 4 and 15 seconds, then generate. Draft at 768p and move to 2K once the shot works.
Specifications
Use Cases
From social content to professional workflows, see how creators and teams are using AI video generation across industries.
Macro product motion with controlled lighting for launches, ads and marketplace listings.
Character-driven shots with natural light and subtle camera movement for editorial and campaign work.
Handheld footage with practical light and real movement for documentary, travel and food storytelling.
Wide natural shots with smooth camera movement for openings, transitions and background plates.
Everything you need to know about MiniMax H3.
Visual media on this page was generated with MiniMax H3 in Vofy Studio and illustrates the creative workflows described. It is not a benchmark or a guarantee of output.