Generate video from image, video and audio references with MiniMax H3 Spicy on the Muapi API. Relaxed-moderation tier at 480p, 768p or 2K. Pay per generation.
MiniMax H3 Spicy Reference to Video is the relaxed-moderation tier of MiniMax H3, generating video with native audio from a prompt plus image, video and optional audio references at 480p, 768p or 2K through the Muapi API.
About this model
MiniMax H3 Spicy Reference to Video is the relaxed-moderation tier of MiniMax H3. A prompt plus image or video references guide subjects, motion, style and scene continuity, with optional reference audio, native audio output, 5-15 seconds per clip at 480p, 768p or 2K. Inputs match the standard model, with adaptive available as an aspect ratio and an optional aigc_watermark flag. Output is priced at $0.058/second at 480p, $0.115/second at 768p and $0.161/second at 2K. It follows Muapi's standard asynchronous request, poll and webhook pattern.
Cost analysis
| Provider | Cost | Notes |
|---|---|---|
| muapiapp | $0.058/sec (480p); $0.115/sec (768p); $0.161/sec (2K) | Output and reference-video rates follow the selected resolution. The first five reference images and audio are free; each additional image is $0.0575. |
| Fal.ai | Not available | MiniMax H3 Spicy is not listed on Fal.ai. |
| Replicate | Not available | MiniMax H3 Spicy is not listed on Replicate. |
Output and reference-video rates follow the selected resolution. The first five reference images and audio are free; each additional image is $0.0575.
MiniMax H3 Spicy is not listed on Fal.ai.
MiniMax H3 Spicy is not listed on Replicate.
** Competitor pricing is estimated based on similar model architectures and usage tiers.
Configuration schema
| Parameter | Type | Description | Default |
|---|---|---|---|
| Prompt | string | Describe the video scene and how the references should be used. | Use the references to create a cinematic product reveal with a slow camera orbit. |
| Reference Images | array | Reference image URLs. Provide at least one image or video reference. | undefined |
| Reference Videos | array | Reference video URLs. Provide at least one image or video reference. | undefined |
| Reference Audio | array | Optional reference audio URLs. Audio cannot be used alone. | undefined |
| Aspect Ratio | Enum (7 options) | - | adaptive |
| Resolution | Enum (3 options) | - | 768p |
| Duration | Enum (11 options) | Output duration in seconds. | 5 |
| AIGC Watermark | boolean | Optional. Add an AIGC watermark to the generated video. | - |
Describe the video scene and how the references should be used.
Use the references to create a cinematic product reveal with a slow camera orbit.Reference image URLs. Provide at least one image or video reference.
undefinedReference video URLs. Provide at least one image or video reference.
undefinedOptional reference audio URLs. Audio cannot be used alone.
undefined-
adaptive-
768pOutput duration in seconds.
5Optional. Add an AIGC watermark to the generated video.
-Developer documentation
Add references: Provide at least one URL in reference_images or reference_videos. Optionally add reference_audios; audio cannot be used alone.
Write the prompt: Describe the scene and how to use the references.
Pick output settings: Choose aspect_ratio (adaptive, 21:9, 16:9, 4:3, 1:1, 3:4, 9:16), resolution (480p, 768p, 2k) and duration (5-15 seconds).
Submit and retrieve: POST the request, then poll /predictions/{request_id}/result or supply a webhook_url.
curl -X POST https://api.muapi.ai/api/v1/minimax-h3-reference-to-video-spicy \
-H "x-api-key: YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"prompt": "Use Image 1 as the character in a slow cinematic orbit shot.",
"reference_images": ["https://example.com/character.png"],
"aspect_ratio": "adaptive",
"resolution": "768p",
"duration": 5
}'
Frequently asked
One or more image or video URLs, with optional audio URLs. Audio cannot be used alone.
Reference audio and the first 5 reference images are free; each additional image costs $0.0575. Reference video seconds are billed at the selected output rate: $0.058/second at 480p, $0.115/second at 768p or $0.161/second at 2K.
Resolution can be 480p, 768p or 2K, and duration can be any value from 5 to 15 seconds.
Failed generations are refunded automatically.