Generate 1080p video from text with the MiniMax H3 Max API on Muapi. Cinematic motion, prompt expansion, pay per generation. MiniMax H3 Max Text to Video creates high-fidelity video up to 1080p from a written prompt through the Muapi API.
MiniMax H3 Max Text to Video creates high-fidelity video up to 1080p from a written prompt through the Muapi API.
Acerca de este modelo
MiniMax H3 Max Text to Video turns a detailed written prompt into a video clip at up to 1080p resolution. It supports 5-15 second durations, six aspect ratios, and an optional prompt-expansion step that rewrites your prompt for stronger cinematic detail before rendering. The model is built for scenes that need accurate camera movement, lighting, and subject action described purely in text, and it slots directly into Muapi's standard asynchronous request/poll/webhook workflow like every other Muapi video model. For image-guided motion instead of a pure text prompt, see MiniMax H3 Max Image to Video; for combining multiple reference assets, see MiniMax H3 Max Reference to Video.
Análisis de costes
| Proveedor | Coste | Notas |
|---|---|---|
| muapiapp | From $0.0325/sec (480p) up to $0.104/sec (1080p) | Launch pricing through Sep 30, 2026 (UTC): From $0.0325/sec (480p) up to $0.104/sec (1080p). From Oct 1, 2026 the standard rate applies: $0.065–$0.208/sec by resolution. Pay-per-generation, no subscription. |
| Replicate | No disponible | MiniMax H3 Max is not listed on Replicate. |
Launch pricing through Sep 30, 2026 (UTC): From $0.0325/sec (480p) up to $0.104/sec (1080p). From Oct 1, 2026 the standard rate applies: $0.065–$0.208/sec by resolution. Pay-per-generation, no subscription.
MiniMax H3 Max is not listed on Replicate.
** Los precios de competidores son estimaciones basadas en arquitecturas de modelos y niveles de uso similares.
Esquema de configuración
| Parámetro | Tipo | Descripción | Por defecto |
|---|---|---|---|
| Prompt | string | vídeo, , . | A cinematic tracking shot of a silver sports car driving through a rain-soaked city at night. |
| Relación de aspecto | Opciones permitidas (6) | - | 16:9 |
| Resolución | Opciones permitidas (3) | Output resolution. 1080p applies latent refinement from a 768p source. | 768p |
| Duración (segundos) | Opciones permitidas (11) | salidaduración () . | 5 |
| Prompt Expansion Mode | Opciones permitidas (3) | Prompt rewriting effort. 'quality' can add up to ~30s of processing. | balanced |
vídeo, , .
A cinematic tracking shot of a silver sports car driving through a rain-soaked city at night.-
16:9Output resolution. 1080p applies latent refinement from a 768p source.
768psalidaduración () .
5Prompt rewriting effort. 'quality' can add up to ~30s of processing.
balancedDocumentación para desarrolladores
Write the prompt: Describe the scene, subject action, camera movement, lighting, and visual style in prompt. Detailed prompts produce more controllable results.
Pick resolution and duration: Set resolution to 480p, 768p, or 1080p, and duration to any value from 5 to 15 seconds. Higher resolution and longer duration cost more.
Set aspect ratio: Choose aspect_ratio from 21:9, 16:9, 4:3, 1:1, 3:4, or 9:16 to match your target platform.
Choose prompt expansion (optional): Set prompt_expansion_mode to disabled, balanced (default), or quality to control how much the model rewrites your prompt before rendering. quality adds up to ~30 seconds of processing.
Submit and retrieve: POST the request, then poll /predictions/{request_id}/result or supply a webhook_url to receive the finished video URL.
curl -X POST https://api.muapi.ai/api/v1/minimax-h3-max-text-to-video \
-H "x-api-key: YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"prompt": "A cinematic tracking shot of a silver sports car driving through a rain-soaked city at night.",
"aspect_ratio": "16:9",
"resolution": "1080p",
"duration": 6
}'
Preguntas frecuentes
It creates a video clip from a written prompt, rendering scene details, subject action, camera direction, and visual style at up to 1080p resolution.
Resolution can be 480p, 768p, or 1080p, and duration can be any value from 5 to 15 seconds.
It controls how much the model rewrites your prompt before generation. `disabled` uses your prompt as-is, `balanced` applies light rewriting, and `quality` applies the most thorough rewrite at the cost of extra processing time.
Poll the request ID returned by the initial POST at `/predictions/{request_id}/result`, or pass a `webhook_url` to be notified when the video is ready.
21:9, 16:9, 4:3, 1:1, 3:4, and 9:16 are all supported, so you can target widescreen, square, or vertical placements.
Yes. Cost scales with both resolution and duration — 1080p output costs more per second than 768p or 480p.