Wan 2.1 T2V 480p LoRA Ultra Fast API Documentation

Wan 2.1 T2V 480p LoRA Ultra Fast API Documentation

Playground

WAN 2.1 T2V 480p delivers ultra-fast text-to-video generation with custom LoRA support for unlimited 480p AI videos. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Wan 2.1 T2V 480p LoRA Ultra Fast is a low-latency text-to-video model designed for rapid iteration. It generates short 480p clips from a single prompt, and supports adding LoRAs to steer style, characters, or motion patterns while keeping throughput high.

Key capabilities

• Text-to-video generation at 480p
• Ultra-fast inference for quick previews and batch exploration
• LoRA support for style/character control (up to 3 LoRAs per run)
• Prompt-driven motion, staging, and shot direction
• Works well for storyboard drafts, social content, and concept tests

Use cases

• Fast storyboard and pre-vis from a director-style prompt
• Stylized “template clips” using LoRAs (brand look, character identity, anime/toy/film looks)
• Social content ideation: generate 10–20 variations quickly, pick the best, upscale later
• LoRA-driven series: consistent visual language across multiple clips for campaigns

Pricing

Inputs

• prompt (required): describe subject, action, scene, camera, and style

• negative_prompt (optional): reduce blur, jitter, distortions, low-quality artifacts

• loras (optional): up to 3 LoRAs, each with:

• path: owner/model-name or a direct.safetensors URL

• scale: LoRA strength (commonly around 0.6–1.0 to start)

Parameters

• duration: clip length (commonly 5s or 10s)
• size: output size preset (e.g., 832×480)
• num_inference_steps: more steps can improve stability/detail at the cost of speed
• guidance_scale: higher values follow the prompt more strongly (can reduce natural motion if too high)
• flow_shift: motion behavior tuning (useful for more/less dynamic motion)
• seed: set for reproducible results (-1 for random)

Prompting tips (T2V)

• Write like a shot list: subject + action + environment + camera + style
• Keep motion explicit: “slow pan left”, “subtle head turn”, “hands gesture while talking”
• For multi-panel or UI-like scenes (e.g., a video call), specify layout and per-panel actions clearly
• If using LoRAs, keep the base prompt simpler and let the LoRA carry most of the style signal

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",
  "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/t2v-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