Minimax H3 Reference To Video LoRA API Documentation

Minimax H3 Reference To Video LoRA API Documentation

Playground

MiniMax H3 Open Weights Reference to Video with custom LoRA support generates coherent 480P / 768P videos from prompts and multimodal references, guided by up to 9 reference images, 3 reference videos, and 3 reference audios, with native stereo audio and flexible reference-based video generation on WaveSpeedAI infrastructure. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Run the open-weights edition of MiniMax H3 on WaveSpeedAI’s own GPU infrastructure. This endpoint is separate from the official minimax/h3 API: same model family, independently hosted, with its own 480p/768p resolutions and per-second pricing.

MiniMax H3 is an omni-modal video model that generates picture and native stereo audio in a single pass. In reference-to-video mode you guide the generation with up to 9 reference images, 3 reference videos, and 3 reference audio tracks, then tell the model — in the prompt — what each reference is for.

How to Use References

Refer to every input in the prompt with an angle-bracket tag: <Picture 1>, <Video 1>, <Audio 1>, and so on. The tags must be written exactly like that, with the brackets — plain text such as Picture 1 is treated as ordinary words, not a reference.

Assign a job to each reference. Explicit assignments work far better than leaving the model to guess:

> “Use the character from <Picture 1>, place them in the setting from <Picture 2>, and match the camera motion of <Video 1>.”

Numbering

• Tags are numbered per type, in the order the inputs are provided: images are <Picture 1>…<Picture 9>, standalone audios are <Audio 1>…, videos are <Video 1>…<Video 3>.
• A reference video’s own soundtrack is used automatically and occupies the earliest <Audio …> slots; any standalone reference_audios you provide are numbered after the video soundtracks. If you only need the video’s audio, refer to it via its <Video …> tag.

Rules

• At least one reference image or video is required. Audio cannot be provided alone.
• reference_videos is available only at 480p output resolution.
• Reference video soundtracks are picked up automatically; total reference-video duration shares a 15-second budget and longer inputs are trimmed fairly.

How to Write a Great Prompt

Write the prompt as a timeline with a schedule, and weave the reference tags into it:

1. Style & references — establish the look and say which reference drives identity, style, motion, or voice.
2. Timeline — timestamped beats: “[0s-3s] <Picture 1> stands on the balcony from <Picture 2> … [3s-6s] turns toward camera and smiles …”.
3. Camera — explicit movement, including when to hold: “slow orbit, no cuts”.
4. Audio — an Audio: line with entrance cues; reference an input voice or track by tag if you want it matched.
5. On-screen text — spell out readable words in quotes and add “no subtitles, do not add other text.”

Parameters

At least one of reference_images, reference_videos, or reference_audios is required.

Specifications

Pricing

Output video is billed per generated second, plus per-reference charges:

Explore the MiniMax H3 Open Weights Family

• MiniMax H3 Text-to-Video Open Weights — Generate video with native stereo audio directly from a text prompt.
• MiniMax H3 Image-to-Video Open Weights — Animate a first-frame image with prompt-guided motion and audio.
• MiniMax H3 Reference-to-Video Open Weights — Generate video from image, video, or audio references with prompt guidance.

Authentication

For authentication details, please refer to the Authentication Guide.

API Endpoints

Submit Task & Query Result

set -euo pipefail

export WAVESPEED_API_KEY="your-api-key"

REQUEST_BODY=$(cat <<'JSON'
{
  "prompt": "A cinematic ocean wave at sunrise, highly detailed",
  "aspect_ratio": "16:9",
  "resolution": "480p",
  "duration": 5,
  "seed": -1
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/minimax-h3/reference-to-video-lora" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}")

TASK=$(printf '%s' "${SUBMIT_RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "${TASK}" | jq -r '.id // empty')
if [ -z "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id
' >&2
  exit 1
fi
RESULT_URL=$(printf '%s' "${TASK}" | jq -r '.urls.get // empty')
if [ -z "${RESULT_URL}" ]; then RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"; fi

# 2. Poll until the prediction finishes.
while true; do
  RESPONSE=$(curl --silent --show-error --fail-with-body \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}")
  RESULT=$(printf '%s' "${RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout) printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    created|processing) sleep 2 ;;
    *) printf 'Unexpected status: %s
' "${STATUS}" >&2; exit 1 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Response Parameters

Result Request Parameters

Result Response Parameters