MiniMax H3
MiniMax H3 is an AI video model that generates 5 to 15-second 2K videos from text, frames, and multimodal media via a web browser.
Open productWhat it does
About MiniMax H3
MiniMax H3 is an AI video model accessible through an online browser workflow. It creates short 2K videos with durations ranging from 5 to 15 seconds using text prompts, frame inputs, or multimodal reference media. Users can guide how generated videos start and end by uploading a starting frame, an ending frame, or both. Additionally, prompt instructions can specify camera movements including pans, zooms, tracking shots, and static shots.
- Availability:
- launched
- Access:
- freemium
- Format:
- software
- Platform:
- web
Pricing at launch: New accounts receive 90 credits upon signup to test video generation. Paid credit packs are non-recurring one-time purchases that do not expire, consuming 9 credits per second for 768P videos and 14 credits per second for 2K videos.
Core capabilities
Multimodal Video Generation
Generates 2K resolution videos lasting between 5 and 15 seconds using text prompts, initial and ending frames, or multimodal reference media.
First and Last Frame Guidance
Keep discovering
Related debuts
Continue with products that solve a similar problem or serve the same kind of maker.
