Skip to content

MiniMax H3 / Text to video

Storyboard a short MiniMax H3 clip around one clear action

A short AI clip needs a small story: a starting state, a visible change and a readable ending. A five-second creative brief is a useful planning exercise, although JollyAI's public H3 duration presets currently differ from the administrator setting used for our five-second demonstration.

Design the action before the prompt

Give the first moment to recognition. Let the central action occupy most of the shot, then leave a small settling period. A light revealing a product, a curtain opening onto a room or a bicycle passing the camera can each fit this structure. Several location changes usually cannot.

Timing language expresses intent; it does not guarantee a frame-accurate schedule. If an edit requires an exact five-second slot, choose a suitable available duration and trim the approved output in an editor. Check the exported file's duration rather than relying only on the requested setting.

Write a three-beat brief

  1. Opening: name the subject and its initial position.
  2. Action: describe one change with a clear direction.
  3. Ending: describe a stable state that gives you an edit point.
  4. Add one camera instruction that supports all three beats.

Example storyboard prompt

PROMPT RECIPE / ADAPT TO YOUR SOURCE

A folded paper lantern rests on a dark tabletop. In one continuous close shot it slowly opens into a glowing round lantern, then remains still for a brief finishing hold. The camera stays locked. Warm light spreads across the table. Sound: soft paper movement and quiet room tone.

Edit for clarity

Watch the result without sound first: the action should still be understandable. Then listen for an audio ending that can be cut cleanly. If the main change happens too late, simplify the opening description or use an earlier trim. Avoid adding a second event to compensate for a weak first event; regenerate the simpler brief instead.

REVIEWED JOLLYAI IMPLEMENTATION

The workflow behind the guide

Written scene → text conditioning → turbo audio/video sampling → separate video and audio decoding → combined video file.

Inputs, presets & access
Limited free access; four-second free preset, with six/eight/ten-second PRO options. Current ordinary pixel budget: 0.4 MP. Use the account controls for current availability.
Implementation note
Local H3 turbo workflow. Six sampling steps; simple scheduler; turbo adapter; video/audio shifts 12/3. Current environment: pruned W4A8 FL2VA weights with a 4B encoder and matching projection. R2V uses a separate reference checkpoint. These settings describe the reviewed JollyAI deployment, not every H3 implementation.

Configuration reviewed September 20, 2026. Your current playground controls and plans page determine availability. Prompt recipes above are suggestions, not guarantees or claimed test results.

Open Text to video ↗

ORIGINAL JOLLYAI GENERATION

A five-second request, inspected as a real file

A fictional green sports shoe on a dark pedestal; the viewpoint shifts as light and mist move. Generated ambience, no dialogue. Selected concept-product example. The shoe remains recognizable, but the viewpoint changes more than a strict lateral slide and some fine details vary. Inspect the full clip; this is not a fidelity guarantee or a matched benchmark.
Request
5 seconds / 0.4 MP
Delivered file
864 × 480 / 24 fps / 5.167 s
Audio
Generated audio stream
Queue wait
22 seconds
Processing interval
115 seconds
End to end
137 seconds

One owner-account observation, September 20, 2026. Processing is the recorded job interval, including pipeline overhead, not isolated sampling time. Other requests and settings can differ. Five seconds was an administrator request; public H3 presets currently use 4/6/8/10 seconds.

Exact demonstration prompt

One continuous studio product film of a fictional unbranded acid-green sports shoe resting on a matte charcoal pedestal. The camera slowly slides from left to right through a modest angle while the shoe remains stationary. A thin veil of pale mist drifts behind the pedestal. A narrow white light sweeps across the fabric texture once, revealing the mesh and sole. Clean black background, sharp silhouette, believable proportions, premium commercial photography, no writing or logos. Sound: a soft airy whoosh and quiet studio ambience. Landscape 16:9.

Sources and useful next steps

The implementation notes describe JollyAI's reviewed local integration. Upstream capabilities and licences may differ from the settings available here.

Found an outdated setting or a reproducible problem? Contact JollyAI with the workflow and job reference. Keep account tokens and private input files out of public reports.