MiniMax H3 Lite API

MiniMax H3 Lite text-to-video API generates 4–15 second clips with sound at 480p or 768p. Write the scene once, then rerun it with one change at a time to see which take works. Its lightweight design suits trying ideas and producing multiple variations. Every request gives you one MP4 in which picture and sound come out together.

Choose a mode by starting material

Sound and dialogue in one pass

Every MiniMax H3 Lite clip comes back with sound. Put spoken lines in double quotes and describe music or ambience in words; both arrive with the clip, so a quick draft needs no separate audio pass.

Six aspect ratios

All six ratios cost the same, so one idea can be tested as a landscape cut and as a vertical cut at no difference in price.

Two price points

480p costs about 40% less per second than 768p, which makes it the cheap way to explore a prompt before rendering the final take.

Storyboards and pitch clips

Turn a written shot list into rough moving previews before anyone commits to a production. Each prompt on MiniMax H3 Lite returns a finished clip with sound, so the room hears the scene as well as seeing it, and a weak idea can be dropped before anything expensive is made.

Prompt drafts at volume

Writers and app builders can try many versions of a scene on MiniMax H3 Lite and keep the one that works, paying per second of output only. Change one thing per run, such as the camera, the outfit or a line of dialogue, and the batch shows which version deserves a sharper render.

Vertical clips with a line of dialogue

A 9:16 scene of 5 to 8 seconds with one quoted line and ambience fits social formats directly from MiniMax H3 Lite, with no editing pass for sound. If you post often, keep one setting fixed and vary only the action and the line, so each 4-second draft at 480p costs $0.10.

# Generate the idempotency key once; reuse it when retrying after a lost response.
export SPICY_IDEMPOTENCY_KEY="${SPICY_IDEMPOTENCY_KEY:-$(uuidgen)}"
curl -X POST https://api.spicyapi.ai/api/v1/jobs/createTask \
  -H "Authorization: Bearer $SPICY_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $SPICY_IDEMPOTENCY_KEY" \
  -d '{
  "model": "minimax/h3-lite/text-to-video",
  "input": {
    "prompt": "A slow aerial shot drifting over a misty pine forest at sunrise",
    "resolution": "768p",
    "aspect_ratio": "16:9",
    "duration_seconds": 5
  }
}'

# Copy data.taskId from the response into TASK_ID, then poll the status.
TASK_ID="TASK_ID_FROM_RESPONSE"
curl "https://api.spicyapi.ai/api/v1/jobs/recordInfo?taskId=$TASK_ID" \
  -H "Authorization: Bearer $SPICY_API_KEY"

See the API reference for authentication, the complete schema, polling, and error handling.