Wan 2.2 I2V LoRA Trainer API Documentation

Wan 2.2 I2V LoRA Trainer API Documentation

Playground

Train custom Wan 2.2 I2V LoRA models 10x faster. Action training, motion training, video efect training. From concept to model in minutes, not hours. Upload a ZIP file containing videos to start!

Features

Wan 2.2 I2V (Image-to-Video) LoRA Trainer is a specialized training service for creating custom LoRA models optimized for image-to-video generation. Train personalized models 10x faster using video datasets to achieve action and video effect generation from static images.

Training Architecture

Built on Wan 2.2’s advanced MoE (Mixture of Experts) architecture, the trainer generates two specialized LoRA models:

• high_noise_lora: Optimized for high-noise denoising timesteps, handling initial motion planning and temporal structure
• low_noise_lora: Optimized for low-noise denoising timesteps, refining motion details and ensuring smooth transitions

This dual-model approach ensures superior image-to-video conversion quality with consistent temporal coherence.

Training Process

1. Video Data Upload: Upload a ZIP file containing your training video sequences
2. Temporal Analysis: The system analyzes motion patterns and temporal relationships
3. Dual Model Training: Simultaneously trains both high_noise_lora and low_noise_lora models
4. Motion Optimization: Fine-tunes models for smooth image-to-video transitions
5. Model Delivery: Receive two specialized LoRA models optimized for I2V generation

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'
{
  "data": "https://github.com/mdn/interactive-examples/archive/refs/heads/main.zip",
  "steps": 100,
  "learning_rate": 0.0002,
  "lora_rank": 32
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/wan-2.2-i2v-lora-trainer" \
  -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