H3 Max Multi Angle turns a single image into a video with precise, keyframe-based control over the camera's orbit, elevation, and distance in 3D space

minimax/h3-max/camera-controls

H3 Max Multi Angle turns a single image into a video with precise, keyframe-based control over the camera's orbit, elevation, and distance in 3D space

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/camera-controls

Schema

Input

Prompt*

Additional Settings

Customize your input with more control.

Result

Idle

What would you like to do next?

H3 Max Camera Controls (Image to Video) API on fal

Input

Prompt

The reference image is one frozen instant. Only the camera moves. Preserve every person and object in exactly the same world position, orientation, shape and pose throughout the shot. Airborne objects remain suspended at the captured height and angle: no wobbling, shaking, spinning, drifting, falling or continued action. Keep faces, hands, clothing, liquids and the background motionless while retaining their natural appearance. Camera parallax is the only source of apparent movement. Follow the supplied camera path smoothly, easing into and out of the move. Start and finish at the original framing and perspective, with unchanged lighting and focal length. No cuts, zoom, morphing or added objects.

Type # to reference inputs.

Resolution

Prompt Expansion Mode*

Image URL*

Hint: Drag and drop image files from your computer, images from web pages, paste from clipboard (Ctrl/Cmd+V), or provide a URL. Accepted file types: jpg, jpeg, png, webp, gif, avif, heic, heif

Camera motion

Start from a preset, then adjust anything below

Drag the camera to place it · scroll on it for distance · drag the background to look around

0:00.0 / 0:06.0

Keyframes

Duration

Keyframe 1: move the camera to change it

Azimuth0°

front view

Elevation0°

Distance1.0

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/0aa9eccc/zSex_5agYoxtKQ6oUdNET_minimax-h3.mp4",
    "content_type": "video/mp4",
    "file_name": "zSex_5agYoxtKQ6oUdNET_minimax-h3.mp4",
    "file_size": 10273057
  },
  "expanded_prompt": "The camera follows the supplied camera trajectory exactly in one continuous shot. Preserve the subject's identity and the scene's spatial consistency throughout the trajectory, with no cuts or additional camera movements.\n\nFor the target video, at 0.00 seconds into the target video, <Picture 1> (from [Shot 1]) is fully referenced.\n\nintegrated_multimodal_description:\n[Shot 1] Based on <Picture 1>, the scene captures a moment of suspended motion at an outdoor recreational facility during the golden hour. A male dancer, dressed in a black-and-white horizontal striped t-shirt, loose black trousers, and white sneakers with a logo, is frozen in a dynamic breakdance freeze known as a \"headspin freeze\" or \"airchair.\" His body is supported entirely by his left hand, which is planted flat and firmly on the light-colored wooden flooring. His right arm is extended upwards, balancing the air. His legs are split horizontally in front of his torso, appearing to float above the ground, while his head tilts back toward the right side of the frame with a focused expression. He wears a black baseball cap.\n\nThe entire frame is a static tableau. The dancer’s pose is rigidly maintained throughout the 6-second duration, acting as a sculpture in time. There is absolutely no movement in his limbs, no swaying of his torso, and no change in the position of his feet or hands relative to the wood. The background remains frozen in place: the chain-link backboards of a tennis court stand upright on the left, the \"TNT\" utility pole and its fixtures on the right remain stationary, and the houses and trees in the distant background show no signs of wind or motion. The sky presents a still, soft gradient of pale yellow and green.\n\nThe movement is purely positional. As the perspective shifts, the spatial relationship between the viewer and the subject changes, creating a strong sense of depth. The parallax effect causes the background elements, such as the net and the houses, to slide laterally relative to the dancer, revealing more of the surrounding suburban environment while the dancer remains the immovable anchor of the composition. The focus stays sharp on the dancer's hand on the floor, emphasizing the tension in his fingers and the detail of the wood grain, while the background slightly softens to maintain that separation.\n\noverall_soundscape:\nAmbience of a quiet outdoor urban space with distant street sounds.\nA single sharp thud of a hand impacting wood at the start.\nFaint, dry wind moving through tree branches in the background.\n\nnon_diegetic_music:\nN/A",
  "timings": {
    "inference": 4.8409632469993085
  }
}

Billing is calculated per second of video generated. Through October 31, 2026, the 35% promotional discount applies: $0.0325 at 480p, $0.052 at 768p, and $0.104 at 1080p per second. From November 1 through November 15, 2026, get 30% off standard rates: $0.035 at 480p, $0.056 at 768p, and $0.112 at 1080p per second. During the November promotion, 5 seconds of video at 768p costs $0.28. From November 16, 2026, standard rates are $0.05, $0.08, and $0.16 per second at 480p, 768p, and 1080p, respectively. Contact your sales representative to discuss permanent discounted pricing.