Generate video with synchronized audio from a prompt and image, video, or audio references with H3 Max Turbo, at 480p, 768p, or 1080p for up to 15 seconds.

minimax/h3-max-turbo/reference-to-video

Generate video with synchronized audio from a prompt and image, video, or audio references with H3 Max Turbo, at 480p, 768p, or 1080p for up to 15 seconds.

Inference

Commercial use

Schema

Playground API Examples

Input

Prompt*

Additional Settings

Customize your input with more control.

Result

Idle

What would you like to do next?

minimax/h3-max-turbo/reference-to-video

Schema

Input

Prompt*

Additional Settings

Customize your input with more control.

Result

Idle

What would you like to do next?

Input

Prompt*

The woman from Image 1 walks barefoot onto the mirror-wet black sand beach from Image 2 at blue hour, then breaks into a slow, sweeping contemporary dance: she spins, arms open, and her long crimson silk coat flares out and ripples in the sea wind, its reflection mirrored in the thin sheet of water. Low fog drifts behind her, the basalt sea stack stands on the horizon. The camera slowly orbits around her at waist height. Sound of gentle waves, wind, and the soft rustle of silk.

Type # to reference inputs.

Reference Image URLs

Hint: Drag and drop files from your computer, images from web pages, paste from clipboard (Ctrl/Cmd+V), or provide a URL.

2 images added

Reference Video URLs

Hint: Drag and drop files from your computer, videos from web pages, paste from clipboard (Ctrl/Cmd+V), or provide a URL.

Reference Audio URLs

Hint: Drag and drop files from your computer, audio from web pages, paste from clipboard (Ctrl/Cmd+V), or provide a URL.

Duration

Resolution

Prompt Expansion Mode*

Aspect Ratio

Additional Settings

Customize your input with more control.

Result

Idle

What would you like to do next?

{
  "video": {
    "url": "https://v3b.fal.media/files/b/0ab0815c/Q5hJ48Lg5TLmCKsQrfsHO_minimax-h3.mp4",
    "content_type": "video/mp4",
    "file_name": "Q5hJ48Lg5TLmCKsQrfsHO_minimax-h3.mp4",
    "file_size": 6866663
  },
  "expanded_prompt": "subject_definitions:\n<Subject 1> is the woman in <Picture 1>, with short dark natural hair, wearing a black tank top, black leggings, and a long crimson silk coat.\n<Subject 2> is the landscape from <Picture 2>, featuring a mirror-wet black sand beach, a thin layer of reflective water, a distant basalt sea stack on the horizon, and a blue hour sky with low drifting fog.\n\nsummary:\n[reference generation] The target video shows <Subject 1> performing a contemporary dance on the beach from <Subject 2> during blue hour.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1]): fully_preserved - her facial features, hair, and the specific crimson silk coat are maintained.\n<Subject 2> (appears in [Shot 1]): fully_preserved - the black sand, sea stack, and blue hour lighting are accurately recreated.\n\ndetailed_description:\nThe video is filmed in a cinematic style during the blue hour, capturing a serene and ethereal atmosphere. [Shot 1] <Subject 1> from <Picture 1> walks barefoot onto the mirror-wet black sand of <Subject 2> from <Picture 2>. The sand is a deep charcoal, reflecting the soft blue and purple of the sky. As she moves, she begins a slow, sweeping contemporary dance. She spins gracefully with her arms open wide; the long crimson silk coat she wears flares out and ripples dynamically in the sea wind. The reflection of the red fabric is clearly visible in the thin sheet of water covering the beach. In the background, low fog drifts across the shoreline, and the silhouette of a basalt sea stack stands against the horizon. The camera performs a slow, smooth orbit around her at waist height, capturing the fluid motion of her body and the flowing coat. The sound consists of the rhythmic, gentle wash of waves hitting the shore, the constant whistle of the wind, and the distinct soft rustle of silk moving.\n\noverall_soundscape:\nAmbient sounds of gentle ocean waves lapping against black sand, a steady sea breeze, and the crisp rustling of silk fabric.\n\nnon_diegetic_music:\nN/A",
  "seed": 1097955473,
  "reference_tokens": 3648,
  "timings": {
    "inference": 5.542492649008636
  }
}

Billing is calculated per second of output video, plus reference inputs beyond the included allowance. Video costs $0.05 per second at 480p, $0.08 per second at 768p, and $0.16 per second at 1080p. A 5-second 768p clip costs $0.40 before additional reference charges. The first 4,096 reference tokens are included; additional reference tokens cost $0.016 per 1K tokens.

Input

Prompt*

Additional Settings

Customize your input with more control.

Result

Idle

What would you like to do next?

Input

Prompt*

Type # to reference inputs.

Reference Image URLs

Hint: Drag and drop files from your computer, images from web pages, paste from clipboard (Ctrl/Cmd+V), or provide a URL.

2 images added

Reference Video URLs

Hint: Drag and drop files from your computer, videos from web pages, paste from clipboard (Ctrl/Cmd+V), or provide a URL.

Reference Audio URLs

Hint: Drag and drop files from your computer, audio from web pages, paste from clipboard (Ctrl/Cmd+V), or provide a URL.

Duration

Resolution

Prompt Expansion Mode*

Aspect Ratio

Additional Settings

Customize your input with more control.

Result

Idle

What would you like to do next?

{
  "video": {
    "url": "https://v3b.fal.media/files/b/0ab0815c/Q5hJ48Lg5TLmCKsQrfsHO_minimax-h3.mp4",
    "content_type": "video/mp4",
    "file_name": "Q5hJ48Lg5TLmCKsQrfsHO_minimax-h3.mp4",
    "file_size": 6866663
  },
  "expanded_prompt": "subject_definitions:\n<Subject 1> is the woman in <Picture 1>, with short dark natural hair, wearing a black tank top, black leggings, and a long crimson silk coat.\n<Subject 2> is the landscape from <Picture 2>, featuring a mirror-wet black sand beach, a thin layer of reflective water, a distant basalt sea stack on the horizon, and a blue hour sky with low drifting fog.\n\nsummary:\n[reference generation] The target video shows <Subject 1> performing a contemporary dance on the beach from <Subject 2> during blue hour.\n\nretention_analysis:\n<Subject 1> (appears in [Shot 1]): fully_preserved - her facial features, hair, and the specific crimson silk coat are maintained.\n<Subject 2> (appears in [Shot 1]): fully_preserved - the black sand, sea stack, and blue hour lighting are accurately recreated.\n\ndetailed_description:\nThe video is filmed in a cinematic style during the blue hour, capturing a serene and ethereal atmosphere. [Shot 1] <Subject 1> from <Picture 1> walks barefoot onto the mirror-wet black sand of <Subject 2> from <Picture 2>. The sand is a deep charcoal, reflecting the soft blue and purple of the sky. As she moves, she begins a slow, sweeping contemporary dance. She spins gracefully with her arms open wide; the long crimson silk coat she wears flares out and ripples dynamically in the sea wind. The reflection of the red fabric is clearly visible in the thin sheet of water covering the beach. In the background, low fog drifts across the shoreline, and the silhouette of a basalt sea stack stands against the horizon. The camera performs a slow, smooth orbit around her at waist height, capturing the fluid motion of her body and the flowing coat. The sound consists of the rhythmic, gentle wash of waves hitting the shore, the constant whistle of the wind, and the distinct soft rustle of silk moving.\n\noverall_soundscape:\nAmbient sounds of gentle ocean waves lapping against black sand, a steady sea breeze, and the crisp rustling of silk fabric.\n\nnon_diegetic_music:\nN/A",
  "seed": 1097955473,
  "reference_tokens": 3648,
  "timings": {
    "inference": 5.542492649008636
  }
}

H3 Max Turbo Reference to Video (Image to Video) API on fal