The model makes the clip. Video API makes it ready to ship.

MiniMax H3 Max Turbo Extend Video

Video → Video·minimax-h3-max·by fal

MiniMax H3 Max Turbo Extend Video adds prompt-guided footage to existing clips, with optional audio references and adjustable resolution for video editing.

Try it now →
API reference
Runtime (p50)
-
Estimated price
$0.025–$0.16 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-extend-video",
    "input": {
        "output": "extended",
        "prompt": "The train reaches the end of the viaduct and curves into a dense autumn forest. The camera keeps following from above as the locomotive disappears into a stone tunnel cut into the mountainside, leaving a trail of white steam glowing in the low golden sun over the treetops. The whistle sounds one last time and echoes across the valley.",
        "duration": 6,
        "video_url": "https://cdn-us.eachlabs.ai/defaults/f14f4251c7ac4a10a8212826e6d4ee33.mp4",
        "resolution": "1080P",
        "aspect_ratio": "auto",
        "enable_prompt_expansion": true
    },
    "webhook_url": ""
}' \
  https://api.eachlabs.ai/v1/prediction/

Related models

4 models
Topaz Apollo AI model preview
Topaz Apollotopaz
Hover Preview Sprites AI model preview
Hover Preview SpritesUnified AI API for hundreds models. Access video, image, and audio generation from leading providers through one integration with eachlabs.
Volume Control AI model preview
Volume ControlUnified AI API for hundreds models. Access video, image, and audio generation from leading providers through one integration with eachlabs.
Genjutsu Motion Transfer AI model preview
Genjutsu Motion TransferHiggsfield
* FAQ

About MiniMax H3 Max Turbo Extend Video

01 / 03

What is MiniMax H3 Max Turbo Extend Video?

MiniMax H3 Max Turbo Extend Video is a video-to-video model that adds prompt-guided footage to an existing clip. You provide a source video and describe what should happen next. It supports one to fifteen seconds of new footage, optional audio references, and adjustable output resolution.