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 / 540P / 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. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

MiniMax H3 Reference-to-Video LoRA Open Weights runs the open-weights edition of MiniMax H3 with LoRA support on WaveSpeedAI. This endpoint is separate from the official minimax/h3 API and uses its own 480p / 768p / 1080p resolution options and per-second pricing.

Why Choose This?

• Open-weights MiniMax H3 workflow
  Use the open-weights edition of MiniMax H3 through WaveSpeedAI-hosted infrastructure.

• Multimodal reference guidance
  Guide the output with reference images, reference videos, and reference audio.

• Native stereo audio
  Generate visuals and synchronized stereo audio together in one pass.

• LoRA support
  Load up to 3 LoRA weights per request, each with its own scale.

• Reference-aware prompting
  Use tags such as <Picture 1>, <Video 1>, and <Audio 1> to assign clear roles to each input.

• Flexible aspect ratios
  Supports landscape, vertical, square, classic, and cinematic aspect ratios.

Parameters

At least one of reference_images or reference_videos is required. reference_audios can guide sound, voice, rhythm, or ambience, but audio cannot be used as the only reference input.

Reference Rules

• At least one reference image or reference video is required.
• Reference audio cannot be provided alone.
• Reference video soundtracks are picked up automatically.
• Total reference video duration is capped at 15 seconds.
• Longer reference video inputs may be trimmed within the 15-second budget.

How to Use

1. Add reference media — Provide reference images or reference videos. Add reference audio when soundtrack, voice, rhythm, or ambience guidance is needed.
2. Write the prompt — Use <Picture 1>, <Video 1>, and <Audio 1> tags to explain how each reference should be used.
3. Choose aspect ratio — Select the output format that matches your target platform or composition.
4. Choose resolution — Use 480p for lower-cost generation, 540p for a mid-price step up, 768p for the model’s native canvas, or 1080p for full-HD output. Reference videos work at either; they are conformed to a 480p pixel budget internally, so they stay affordable at 768p.
5. Set duration — Choose a duration from 3 to 15 seconds.
6. Add LoRAs optional — Provide up to 3 LoRA files with individual scale values.
7. Set seed optional — Use a fixed seed when reproducibility is needed.
8. Submit — Generate the final MP4 with native stereo audio.

Pricing

Pricing is based on generated video duration, reference video duration, selected resolution, and the number of reference images and audios.

Output duration and total reference video duration are each capped at 15 seconds for pricing. Reference video duration is rounded up to the next whole second.

Example: a 10s 480p output with 2 reference images and a 5s reference video costs 10 × $0.06 + 2 × $0.02 + 5 × $0.06 = $0.94.

Best Use Cases

• Reference-guided video generation — Use images, videos, and audio to guide identity, setting, motion, or sound.
• Character-consistent scenes — Generate videos that follow character or style references.
• Motion transfer and camera guidance — Use reference videos to guide motion, pacing, or camera rhythm.
• Audio-guided generation — Use reference audio to guide music, voice, ambience, or sound direction.
• LoRA-guided creative workflows — Combine references with LoRA weights for stronger style or subject control.
• Social and cinematic clips — Generate short videos for vertical, landscape, square, or cinematic layouts.

Pro Tips

• Assign a clear job to every reference in the prompt.
• Use exact tags such as <Picture 1>, <Video 1>, and <Audio 1>.
• Write prompts as a timeline with timestamped beats.
• Add an Audio: line to control soundtrack, ambience, voice, and effects.
• Use reference videos when motion, gesture, pacing, or camera behavior matters.
• Use reference images when identity, subject appearance, setting, or style matters.
• Use reference audio when rhythm, mood, voice, or sound design matters.
• Use a fixed seed when comparing reference combinations or LoRA scale changes.
• Keep LoRA combinations focused; too many strong LoRAs can reduce control.

Related Models

• 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.

Reference

• How to train your LoRA?
• How to use your LoRA?

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
}
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 "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id
' >&2
  exit 1
fi
RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"

# 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|deleted) printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    *) sleep 2 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Response Parameters

Result Request Parameters

Result Response Parameters