MiniMax H3 Spicy Image to Video: Image-to-Video
Animate still images with MiniMax H3 Spicy Image to Video. Create bold motion with native stereo audio at 480p or 768p. Try free.
Animate still images with MiniMax H3 Spicy Image to Video. Create bold motion with native stereo audio at 480p or 768p. Try free.
About this model
MiniMax H3 Spicy Image to Video animates a supplied first-frame image into a short, high-motion video with native stereo audio. The Spicy variant is designed for bolder movement, expressive action, and more adventurous visual direction while keeping the source image's subject and composition as the starting point. Add an optional last_image to guide the final frame, describe camera movement and mood in prompt, choose a 480p or 768p output, and set a 3–15 second duration. Use it for stylized product shots, character motion, social clips, and creative experiments. For the standard open-weights image workflow, compare MiniMax H3 Open Image to Video; for text-only scenes, see MiniMax H3 Open Text to Video.
Cost analysis
| Provider | Cost | Notes |
|---|---|---|
| muapiapp | $0.058/秒 (480p) or $0.115/秒 (768p) | Pay per generated second with flexible 480p/768p tiers and 无需订阅. Rates are rounded up to maintain at least a 30% gross margin over the listed model rates. |
| Fal.ai | Inte tillgänglig | This model variant is 未在 Fal.ai. |
| Replicate | Inte tillgänglig | This model variant is 未在 Replicate. |
Pay per generated second with flexible 480p/768p tiers and 无需订阅. Rates are rounded up to maintain at least a 30% gross margin over the listed model rates.
This model variant is 未在 Fal.ai.
This model variant is 未在 Replicate.
** Competitor pricing is estimated based on similar model architectures and usage tiers.
Configuration schema
| Parameter | Type | Description | Default |
|---|---|---|---|
| Bild-URL | string | Required starting image URL. The output canvas follows this image's aspect ratio. | https://cdn.muapi.ai/assets/image_download_11.avif |
| Prompt (Instruktion) | string | Optional description of the scene, action, camera movement, lighting, and mood. | A cinematic ocean wave at sunrise, highly detailed |
| Last Image URL | string | Optional final-frame image URL for interpolation. | - |
| Upplösning | Enum (2 options) | Output video resolution. | 480p |
| Längd | Enum (13 options) | Output video duration in seconds. | 5 |
| Slumpmässigt seed-värde | integer | Random seed. Negative values use a random seed. | -1 |
Required starting image URL. The output canvas follows this image's aspect ratio.
https://cdn.muapi.ai/assets/image_download_11.avifOptional description of the scene, action, camera movement, lighting, and mood.
A cinematic ocean wave at sunrise, highly detailedOptional final-frame image URL for interpolation.
-Output video resolution.
480pOutput video duration in seconds.
5Random seed. Negative values use a random seed.
-1Developer documentation
image_url; the output canvas follows its aspect ratio.prompt for action, camera movement, lighting, mood, and scene direction. Leave it out for model-directed motion.last_image when you want interpolation toward a target final frame.resolution (480p or 768p) and duration (3 to 15 seconds). Use seed when you need repeatable experiments./api/v1/minimax-h3-image-to-video-spicy, then poll the returned request ID or use a webhook.curl -X POST "https://api.muapi.ai/api/v1/minimax-h3-image-to-video-spicy" \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"image_url": "https://example.com/first-frame.jpg",
"prompt": "A cinematic ocean wave at sunrise, highly detailed",
"resolution": "480p",
"duration": 5,
"seed": -1
}'
Frequently asked
It is an image-to-video model that turns a starting image into a bold, expressive video with native stereo audio and optional final-frame guidance.
No. `image_url` is required; `prompt` is optional and can describe the desired action, camera movement, lighting, or mood.
The model supports 480p and 768p output with durations from 3 to 15 seconds.
Yes. Pass `last_image` with `image_url` to guide interpolation toward a target ending frame.
Pricing is calculated per second of generated video: $0.058/sec at 480p and $0.115/sec at 768p. Rates are rounded up to maintain at least a 30% gross margin over the listed model rates.