Context
7000
Input
text, image, audio, video
Output
video
Tool calling
Not listed
About this model
MiniMax H3 is a general-purpose, omni-modal generative system that supports unified understanding of multimodal contexts composed of text, images, video, and audio. It can generate video with native stereo audio at resolutions up to 2K and durations of up to 15 seconds. It is released with open weights under the MiniMax H3 Community License.
Input modalities
text
Accepted as model input
image
Accepted as model input
audio
Accepted as model input
video
Accepted as model input
Providers
Available routing options for this model through OneInfer.
Available
Video generation
$0.0800 / second
Input video
$0.0800 / second
Routing
OneInfer optimized
Pricing
Current OneInfer pricing for this model.
| Usage | Price |
|---|---|
| Video generation768P · No audio · Fixed camera · default | $0.0800 / second |
| Input video768P · No audio · Fixed camera · default | $0.0800 / second |
| Additional input image5 input images included · 768P · No audio · Fixed camera · default | $0.040 / image |
| Video generation2K · No audio · Fixed camera · default | $0.1300 / second |
| Input video2K · No audio · Fixed camera · default | $0.1300 / second |
| Additional input image5 input images included · 2K · No audio · Fixed camera · default | $0.040 / image |
Performance
Published evaluation results associated with this model.
No benchmark data is listed for this model.
API example
curl -X POST https://api.oneinfer.ai/v1/ula/generate-video \
-H "Authorization: Bearer YOUR_JWT_TOKEN" \
-H "Content-Type: application/json" \
-d '{
"provider": "minimax",
"model": "MiniMaxAI/MiniMax-H3",
"prompt": "A red fox running through a snowy forest",
"resolution": "768P",
"aspect_ratio": "16:9",
"duration": 5,
"generate_audio": false,
"camera_fixed": true,
"service_tier": "default"
}'