Wan 2.1 I2V 480p LoRA Ultra Fast API Documentation

Wan 2.1 I2V 480p LoRA Ultra Fast API Documentation

Playground

Wan 2.1 i2v 480p Ultra-Fast generates unlimited image-to-video content at 480p, supporting custom LoRAs for style personalization. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Wan 2.1 I2V 480p LoRA Ultra Fast is a fast, cost-efficient image-to-video model that animates a single reference image into a short clip guided by your prompt. It supports LoRA injection so you can enforce a specific motion/style/character look while keeping generation lightweight—great for rapid iteration, previews, and high-volume variants.

Key capabilities

• Image-to-video (I2V) generation at 480p
• Prompt-driven motion with coherent, image-anchored composition
• LoRA support for consistent style, motion patterns, or character look
• Tunable quality/speed via inference steps
• Great for quick drafts, storyboards, and large-scale A/B testing

Use cases

• Quick animation passes for key art, avatars, and character cards
• Motion experiments (e.g., 360° rotation, walk cycles, subtle facial motion) using LoRAs
• Fast concept validation for ads and social loops
• Batch generation of multiple motion variants from the same image
• Previews before re-running at higher resolution (720p/1080p)

Pricing

Inputs

• image (required): reference image that anchors subject and composition
• prompt (required): action + camera + motion intent
• loras (optional): LoRA items to apply (path + scale)

Parameters

• duration: clip length (commonly 5s or 10s)
• num_inference_steps: sampling steps (higher can improve motion coherence)
• guidance_scale: prompt adherence strength
• flow_shift: motion behavior tuning
• negative_prompt: what to avoid (blur, jitter, artifacts, low quality)
• seed: set for reproducible results (-1 for random)
• size: output size preset (e.g., 832×480)
• loras[].path: LoRA identifier (e.g., creator/name) or a safetensors URL
• loras[].scale: LoRA strength (start around 0.8–1.0)

Prompting guide (I2V)

Keep the prompt like a director note:

• Subject: what’s on screen
• Motion: what changes over time (spin, turn, step forward, hair movement)
• Camera: locked-off / slow push-in / orbit / handheld, etc.
• Constraints: “smooth motion”, “no camera shake”, “keep character consistent”

Example prompt

Felt doll standing on a grassy field. The doll slowly rotates for a full 360-degree turn with smooth motion, stable framing, no jitter. Soft daylight, shallow depth of field, gentle background bokeh.

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",
  "image": "https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg",
  "size": "832*480",
  "num_inference_steps": 30,
  "duration": 5,
  "guidance_scale": 5,
  "flow_shift": 3
}
JSON
)

# 1. Submit the prediction once. Never retry this request automatically:
# a retry after an unclear failure can create a second, billed task.
CHECK='The task may already exist. Check History or the predictions list before submitting again.'
if ! SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  --connect-timeout 10 --max-time 60 \
  -X POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/wan-2.1/i2v-480p-lora-ultra-fast" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}"); then
  printf 'Submit failed. %s\n' "${CHECK}" >&2
  exit 1
fi

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

# 2. Poll until the prediction reaches a final status.
BODY_FILE=$(mktemp)
trap 'rm -f "${BODY_FILE}"' EXIT
while true; do
  # Transient errors (429, 5xx) are retried, honoring Retry-After.
  HTTP_CODE=$(curl --silent --show-error --fail-with-body \
    --connect-timeout 10 --max-time 60 --retry 5 --retry-delay 2 \
    --output "${BODY_FILE}" --write-out '%{http_code}' \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}" || true)
  case "${HTTP_CODE}" in
    2??) ;;
    000|429|5??)  # dropped connection or temporary error: keep polling
      printf 'Task %s: HTTP %s, polling again in 10s\n' "${PREDICTION_ID}" "${HTTP_CODE}" >&2
      sleep 10; continue ;;
    *) printf 'Task %s: HTTP %s\n' "${PREDICTION_ID}" "${HTTP_CODE}" >&2
      cat "${BODY_FILE}" >&2; exit 1 ;;
  esac
  RESULT=$(jq 'if type == "object" and has("data") then .data else . end' "${BODY_FILE}")
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout|deleted)
      printf 'Task %s ended with status %s\n' "${PREDICTION_ID}" "${STATUS}" >&2
      printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    *) sleep 2 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Every model also accepts the global request options enable_sync_mode and enable_base64_output; see their pages for details.

Response Parameters

Result Request Parameters

Result Response Parameters