MiniMax | H3 | Max | Turbo | Image to Video

Video·minimax-h3-max·by fal

H3 Max Turbo Image-to-Video creates short AI video clips from prompts with optional first and last frames for faster cinematic motion.

Runtime (p50)
5m
Estimated price
$0.00625–$0.01 per second, depending on resolution
Call the API
prediction.sh
sh
curl -X POST \
  -H "Authorization: Bearer $EACHLABS_API_KEY" \
  -H "Content-Type: application/json" \
  --data '{
    "model": "minimax-h3-max-turbo-image-to-video",
    "version": "0.0.1",
    "input": {
        "prompt": "A 12 second fast paced fashion campaign film. Energetic, confident, rhythmic cuts that land on the beat. Same bag, same tan leather, same gold hardware in every shot.\n\n0 to 3s: Start on the product shot. The bag rotates quickly to the right on the cream background, light sweeping across the grain, then snaps to a stop facing forward.\n\n3 to 6s: Hard cut to a woman walking away from camera down a sunlit city street, filmed from behind at waist height. Only her lower body and the bag are in frame, face never visible. She wears a tailored cream coat. The bag swings at her side with each step, catching the sunlight. The camera tracks steadily behind her at a fixed distance.\n\n6 to 9s: Hard cut to a tight close up of the bag alone on her shoulder, filmed from the side. Only the leather, the gold hardware and the shoulder of her coat are in frame. No hands, no face. Sunlight moves across the grain as she walks.\n\n9 to 12s: Hard cut back to the clean cream background. The bag lands into frame and settles, front facing, exactly as in the opening shot. The camera holds still for the final beat with clean empty space above the bag.\n\nKeep the bag proportions, color and hardware identical to the source image in every shot. No visible face, no visible hands in any shot. Quick punchy camera moves, real time speed, no slow motion. Bright and premium, high contrast, warm daylight on the street shots.",
        "duration": 12,
        "image_url": "https://cdn-us.eachlabs.ai/defaults/f8790a15a8e8472299cac6b1dcf4e504.png",
        "resolution": "768P",
        "enable_safety_checker": true,
        "prompt_expansion_mode": "quality"
    },
    "webhook_url": ""
}' \
  https://api.eachlabs.ai/v1/prediction/

Related models

4 models
* FAQ

About MiniMax | H3 | Max | Turbo | Image to Video

01 / 03

What is H3 Max Turbo Image-to-Video?

H3 Max Turbo Image-to-Video is a MiniMax video model on fal that generates short videos from text prompts. It can also use an optional first frame and last frame to guide the start and end of the clip.