MiniMax H3 Highspeed
minimax/h3-highspeedMiniMax H3 Highspeed creates 4–15 second videos at 768p with native audio. Generate from text, animate a starting image with an optional end frame, or combine up to nine image, three video, and three audio references. Reference clips must be 2–15 seconds each, totaling at most 15 seconds per media type. Audio references require an image or video. Frame guidance and reference inputs cannot be combined.
Pricing
Includes our margin. Charged per request from your balance.
- Minimum charge
- $0.45
Output duration and resolution determine pricing. Reference media may add charges. See model settings and the generation quote.
Details
- Provider
- MiniMax
- Added
- Oct 24, 2026
Options
Parameters this model accepts.
| Parameter | Values | Default |
|---|---|---|
mode | auto, text-to-video, image-to-video, reference-to-video Default | auto |
duration | 4–15 seconds Default | 5 |
imageUrlOptional start frame for image-to-video. Public HTTP(S) image URL or base64 image data URI (data:image/...;base64,...). Raw base64 without the data URI prefix is not supported. Omit start and end images for text-to-video. Images must be 256–5760 pixels per side, aspect ratio 0.4–2.5. | Text | — |
resolution | 768p Default | 768p |
aspect_ratioImage-to-video follows the starting image. Text and reference generation require a concrete ratio. | 21:9, 16:9, 4:3, 1:1, 3:4, 9:16 Default | 16:9 |
lastImageUrlOptional end frame for image-to-video. Requires a start image. Public HTTP(S) image URL or base64 image data URI (data:image/...;base64,...). Raw base64 without the data URI prefix is not supported. End frames cannot be combined with reference media or reference-to-video mode. | Text | — |
reference_audiosUp to three WAV/MP3 reference audio clips, at most 15 MB per file, 2–15 seconds each, at most 15 seconds combined. Requires an image or video reference. No input-audio charge. | Text | — |
reference_imagesUp to nine image URLs, as an array, JSON array or one per line. Starting images count as a reference in reference mode. Images must be 256–5760 pixels per side, aspect ratio 0.4–2.5. | Text | — |
reference_videosUp to three MP4/MOV reference videos, at most 50 MB per file, 256–5760 pixels per side, aspect ratio 0.4–2.5 and 23.976–60 FPS. Each clip must be 2–15 seconds, at most 15 seconds combined. Duration is verified before billing. | Text | — |
audioAttachmentIdAuthorized media attachment ID as an alternative to a media URL. | Text | — |
imageAttachmentIdAuthorized media attachment ID as an alternative to a media URL. | Text | — |
videoAttachmentIdAuthorized media attachment ID as an alternative to a media URL. | Text | — |
endImageAttachmentIdAuthorized media attachment ID as an alternative to a media URL. | Text | — |