MiniMax · Video
Video from text, first and last frames, and image, video or audio references.
Maker: MiniMax
4–15s · 768p/2K
text→video · photo→video · audioMiniMax H3 makes 4–15 second clips at 768p or 2K with native stereo sound. Start from text, set first and last frames, or add references: up to 9 images, 3 videos and 3 audio files. Audio needs a visual input.
| Type | Video generation (text-to-video, image-to-video) |
|---|---|
| Animate a photo | Yes |
| Input frames | first and last frames |
| References | up to 9 images, 3 videos and 3 audio files; audio needs a visual input |
| Audio | yes, native |
| Clip length | 4–15s |
| Resolution | 768p, 2K |
| Prompt length | 7000 characters |
| Provider model | MiniMax H3 |
| Released | 2026-07-31 |
Video from text, first and last frames, and image, video or audio references. It is a video model by MiniMax (MiniMax H3), available on Mixer AI pay-as-you-go — the price is shown before generation.
Pay as you go, no plans — the price is shown before generation. The exact price is shown before you run it.
Yes — upload a photo as a frame or reference and the model turns it into video. Text-to-video also works.
No. Mixer AI is pay-as-you-go: you top up a balance in coins and spend it only on the generations you want. Available on the site and in the Telegram bot @addbeer_bot.
Mixer AI
Open in Telegram
