MiniMax-H3
MiniMax-H3 is MiniMax's open general-purpose multimodal generation model: it understands a unified context of text, images, video and audio and generates video with native stereo sound at up to 2K resolution and 15 seconds, including multiple shots in one generation.
text + image + audio + video → audio + video · made by MiniMax
Sold by 3 ways
| Seller | Lane | Rate |
|---|---|---|
| MiniMax api | 2k | $0.04per image$7.8per minuteplatform.minimax.io · read 2026-08-25 |
| MiniMax api | 768p | $0.04per image$4.8per minuteplatform.minimax.io · read 2026-08-25 |
| Featherless AI aggregator | standard | $0.04per Mtok in$0.08per Mtok outapi.featherless.ai · read 2026-09-20 |
About
MiniMax-H3 — an audio and video model from MiniMax, sold by 2 companies from $0.04 in and $0.08 out per million tokens.
It takes text, images, audio and video and returns audio and video, with a context window of 2,048 tokens. It was published in July 2026. The catalogue files it under video. Two companies sell it. The cheapest is $0.04 in and $0.08 out per million tokens at Featherless AI. Beside the standard rate there is a separately routed lane.
Every current figure
- Maker
- MiniMax
- Register
- model
- Takes
- text + image + audio + video
- Returns
- audio + video
- Context
- 2,048 tokens
- Published
- July 2026
- Licence
- not read
- Sellers
- 2
- Price
- $0.04 in and $0.08 out per million tokens — Featherless AI
Known as 2 names
MiniMax H3mizzaay/h3