muapi Logomuapi Logo
muapi Logomuapi Logo

MiniMax H3 Max Reference to Video API: Obraz na wideo

Create video from image, video, and audio references with MiniMax H3 Max on Muapi. Up to 1080p, pay per generation. MiniMax H3 Max Reference to Video generates a video clip guided by a prompt plus up to 9 reference images, 3 videos, and 3 audio clips, with no more than 12 reference files combined. Reference the assets by name in your prompt (for example "Image 1" or "Video 1") to tell the model how each one should influence the result — a character's appearance, a background style, a motion pattern, or a soundtrack. Output resolution runs up to 1080p across durations of 5-15 seconds, with an adaptive aspect ratio option that follows the references automatically. It follows Muapi's standard asynchronous request/poll/webhook pattern. For a single starting image without multi-reference blending, see MiniMax H3 Max Image to Video; for a pure text prompt with no references, see MiniMax H3 Max Text to Video.

Interaktywne sterowanie modelem

Create video from image, video, and audio references with MiniMax H3 Max on Muapi. Up to 1080p, pay per generation.

📝

Przegląd

O tym modelu

MiniMax H3 Max Reference to Video generates a video clip guided by a prompt plus up to 9 reference images, 3 videos, and 3 audio clips, with no more than 12 reference files combined. Reference the assets by name in your prompt (for example "Image 1" or "Video 1") to tell the model how each one should influence the result — a character's appearance, a background style, a motion pattern, or a soundtrack. Output resolution runs up to 1080p across durations of 5-15 seconds, with an adaptive aspect ratio option that follows the references automatically. It follows Muapi's standard asynchronous request/poll/webhook pattern. For a single starting image without multi-reference blending, see MiniMax H3 Max Image to Video; for a pure text prompt with no references, see MiniMax H3 Max Text to Video.

1Character consistency: Use multiple reference images of the same character to keep their appearance consistent across a new scene.
2Style transfer: Blend a reference video's motion or camera style with new subject matter.
3Music-driven edits: Layer a reference audio track alongside visual references to guide pacing and mood.
4Product reveals: Combine product reference images with a motion reference video for a polished reveal shot.
💰

Ceny i wartość

Analiza kosztów

muapiappFrom $0.065/sec (480p) up to $0.208/sec (1080p), plus reference item fees

Koszt wyjściowy za sekundę zależy od rozdzielczości; pierwsze 4 elementy referencyjne są wliczone w cenę, a każdy kolejny obraz, film lub plik audio wiąże się z niewielką opłatą ryczałtową. Płatność za generację, bez abonamentu.

ReplicateNiedostępne

MiniMax H3 Max nie jest dostępny w ofercie Replicate.

** Ceny konkurencji są szacowane na podstawie podobnych architektur modeli i poziomów użytkowania.

⚙️

Szczegóły techniczne

Schemat konfiguracji

podpowiedźstring

Describe the video scene and how the references should be used (e.g. 'Image 1', 'Video 1').

Wartość domyślnaUse the references to create a cinematic product reveal with a slow camera orbit.
Obrazy referencyjnearray

Reference image URLs (up to 9). Provide at least one image or video reference. Max 12 combined reference files.

Wartość domyślnaundefined
Filmy referencyjnearray

Reference video URLs (up to 3; 2-15s each, 15s combined max). Provide at least one image or video reference.

Wartość domyślnaundefined
Referencyjny dźwiękarray

Optional reference audio URLs (up to 3; 2-15s each, 15s combined max). Audio cannot be used alone.

Wartość domyślnaundefined
współczynnik proporcjiWyliczenie (opcje 7)

-

Wartość domyślnaadaptive
RozdzielczośćWyliczenie (opcje 3)

-

Wartość domyślna768p
Czas trwania (sekundy)Wyliczenie (opcje 11)

czas trwania produkcji () .

Wartość domyślna5
Prompt Expansion ModeWyliczenie (opcje 3)

Prompt rewriting effort. 'quality' can add up to ~30s of processing.

Wartość domyślnabalanced
📖

Przewodnik wdrożeniowy

Dokumentacja deweloperska

How to Use MiniMax H3 Max Reference to Video

  1. Add references: Provide reference_images (up to 9) and/or reference_videos (up to 3) URLs; reference_audios (up to 3) is optional and cannot be used alone. The combined total across all three fields cannot exceed 12 files, and at least one image or video is required.

  2. Write the prompt: Describe the scene and explicitly reference your assets by name, e.g. "Use Image 1's character walking through the scene from Video 1."

  3. Pick resolution and duration: Set resolution to 480p, 768p, or 1080p, and duration to any value from 5 to 15 seconds.

  4. Set aspect ratio: Leave aspect_ratio as adaptive to follow the references automatically, or choose a fixed ratio.

  5. 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-reference-to-video \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "Use the references to create a cinematic product reveal with a slow camera orbit.",
    "reference_images": ["https://example.com/product.jpg"],
    "resolution": "1080p",
    "duration": 6
  }'
❓

Często zadawane pytania

Często zadawane pytania

What does MiniMax H3 Max Reference to Video generate?

It generates a video guided by a prompt plus one or more reference images, videos, and optional audio, blending their appearance, motion, or sound into the new clip.

How many reference files can I use?

Up to 12 combined reference images and videos, plus optional reference audio tracks up to 15 seconds combined.

Can I use reference audio alone?

No, reference audio must be paired with at least one reference image or video.

What resolutions and durations are supported?

Resolution can be 480p, 768p, or 1080p, and duration can be any value from 5 to 15 seconds.

How does the adaptive aspect ratio work?

When `aspect_ratio` is left as `adaptive`, the output follows the aspect ratio implied by your reference assets instead of a fixed value.

How do I retrieve the finished video?

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.