AI video prompts for MiniMax H3 (Hailuo 3.0), which renders 2K video with stereo audio in one pass. Build brand films, ad spots and shot-level edits from a single prompt.
2026.07.31
「MiniMax H3」を思いどおりに作りましょう自分だけの「MiniMax H3」を作る
ふだんの言葉で伝えるだけで、Caratが最適な画像や動画を生成します。
毎日100クレジット無料で今すぐ始められます
MiniMax H3, also known as Hailuo 3.0, is the video generation model MiniMax released on July 31, 2026. It takes text, images, video and audio together as input and composes them into a single clip. Every render comes out at 2K resolution, 24fps, with stereo audio included.
MiniMax H3 accepts up to 9 images, 3 video clips and 3 audio clips as references in one request. Because you address them by position in the prompt (Image 1, Video 1, Audio 1), you can pull the character from one reference, the camera move from another, and the voice from a third. Generation and editing live in the same model, so shot replacement and relighting are prompt instructions rather than a separate pipeline.
Copying prompts is free. Generating a video uses credits.
They are the same model. Hailuo 3.0 is the product name; MiniMax-H3 is the model and API identifier.
5 to 15 seconds, at 2K resolution and 24fps.
Yes. Every render includes stereo audio, and you can direct dialogue, sound effects and music in the prompt.
Up to 9 images, 3 video clips and 3 audio clips, 12 files in total. You address them in the prompt as Image 1, Video 1 and so on.
Yes. Character and object replacement, background swaps, relighting, and dialogue or voice replacement are all prompt instructions.
21:9, 16:9, 4:3, 1:1, 3:4 and 9:16. If you supply a first-frame image, the output follows that image's ratio.