Unified MiniMax H3 video generation, editing, and extension with native stereo audio. Generate 2K clips from text, images, or multimodal references, edit source videos with optional image/audio guidance, or append seamless 480p/768p continuations.
Added Jul 31, 2026
Approx. Price
$0.120 per video
Model Type
both
Settings
Generation controls available for this model.
Output Format
Default Duration
5
13 duration options
Aspect Ratio
Default
auto
Options (8)
Auto, Cinematic (21:9), Landscape (16:9), Classic (4:3) +4 more
Applies to text-to-video and reference-to-video. Auto lets the model choose.
Duration
Default
5
Options (13)
3 seconds, 4 seconds, 5 seconds, 6 seconds +9 more
Output or appended duration. Edit/extend support 3–15 seconds; generation supports 5–15 seconds.
Edit/Extend Resolution
Default
480p
Options (2)
480p, 768p
Applies to video edit and video extend. Other MiniMax H3 modes generate at 2K.
Generate Edit Audio
Default
Yes
Applies to video edit. Disable to preserve the source video audio.
Mode
Default
auto
Options (4)
Auto-detect, Reference to Video, Edit Video, Extend Video
Auto-detect uses uploads for 2K generation. Video Edit rewrites one source video; Video Extend appends a continuation to one source video.
Seed
Default
-1
Applies to video edit and video extend; use -1 for a random seed.
Benchmarks
Benchmarks
Human preference benchmarks sourced from Artificial Analysis.
Text to Video
#3 / 81
ELO
1302.0
Appearances
5,669
95% CI
-10/10
Image to Video
#3 / 75
ELO
1347.0
Appearances
5,584
95% CI
-11/11
Release Date 2026-07 · Matched as MiniMax H3
Artificial Analysis APIExamples
Loading examples…
Related video models
Compare MiniMax H3 with similar models from the same provider or model family.
MiniMax H3 Max Turbo
minimax/h3-max-turboMiniMax H3 Max Turbo brings H3 Max prompt understanding and aesthetics to video generation at roughly twice the speed and half the cost while targeting 97% of its quality. Create 5–15 second 480p or 768p clips from text or a starting image, with optional first/last-frame transitions.
MiniMax H3 Spicy Image-to-Video
wavespeed-ai/minimax-h3/image-to-video-spicyOpen-weights MiniMax H3 image-to-video generation with expressive unrestricted motion, native stereo audio, optional last-frame control, 3–15 second clips, and 480p or 768p output.
MiniMax H3 Max
minimax/h3-maxMiniMax H3 Max generates 5–15 second videos at 480p or 768p from a text prompt or a first-frame image, with optional first/last-frame transitions.
MiniMax Hailuo 2.3 Pro
minimax-hailuo-23-proMiniMax Hailuo 2.3 Pro delivers cinematic 1080p 5-second clips with advanced physics, prompt fidelity, and character consistency. Supports both pure text prompts and image-conditioned motion.
MiniMax Hailuo 2.3 Standard
minimax-hailuo-23-standardMiniMax Hailuo 2.3 Standard produces 1080p cinematic clips with realistic motion and smooth scene transitions from text prompts or reference images. Choose 6 or 10 second runs for quick drafts or extended shots.
MiniMax Hailuo 02
minimax-hailuo-02MiniMax Hailuo-02 Advanced video generation model with 768p resolution. Creates high-quality videos from text prompts or images. Supports both text-to-video and image-to-video generation.