Longcat Avatar 1.5 API Documentation

Longcat Avatar 1.5 API Documentation

Playground

LongCat Avatar 1.5 is the upgraded LongCat Avatar with sharper lip sync and faster generation. Converts one photo + audio into audio-driven talking or singing avatar videos (Image-to-Video), capped at 64 seconds per clip, 720p tier $0.40/5s. Ready-to-use REST API, no coldstarts, affordable pricing.

Features

LongCat Avatar 1.5 is an upgraded audio-driven avatar video model that turns a single photo into a realistic speaking or singing video. It delivers sharper lip sync, faster generation, natural head and body motion, and strong identity preservation, making it suitable for talking portraits, virtual presenters, short-form content, and avatar-driven storytelling.

Why Choose This?

• Sharper lip synchronization Aligns mouth movement closely with the input audio for more natural speech and singing performance.

• Natural full-body coherence Goes beyond lip motion to preserve believable head movement, facial expression, and body posture.

• Strong identity preservation Maintains facial identity and overall visual consistency across frames.

• Photo-to-avatar video Turns a static image into a lively speaking or singing performance.

• Improved generation speed Faster than earlier LongCat avatar workflows while keeping strong realism.

• Production-ready workflow Suitable for presenter videos, talking portraits, creator content, and personalized avatar clips.

Parameters

How to Use

1. Upload the audio — provide the voice or singing track to drive the avatar.
2. Upload the image — add the portrait photo of the person you want to animate.
3. Add a prompt — guide expression, pose, or visual style if needed.
4. Choose resolution — select 480p or 720p depending on quality and budget needs.
5. Set a seed (optional) — use a fixed value for more reproducible generations.
6. Submit — run the model and download the generated avatar video.

Example Prompt

Natural speaking performance with subtle head movement, calm expression, realistic lip sync, and stable identity.

Pricing

> Billing duration: Input media duration is rounded up to whole seconds. The minimum billed duration is 3 seconds; shorter inputs are billed as 3 seconds. Existing per-5-second rate tables remain unchanged.

Pricing depends on output length and resolution.

Billing Rules

• 480p costs $0.20 per 5 seconds
• 720p costs $0.40 per 5 seconds
• Standard rate at 480p is $0.04 per second
• 720p is 2× the standard rate, or $0.08 per second
• All videos are billed for a minimum of 3 seconds
• Billing is capped at 64 seconds
• Audio longer than 64 seconds is automatically trimmed

Best Use Cases

• Talking portrait videos — Animate a still portrait into a speaking clip.
• Virtual presenters — Create avatar-led explainers, intros, and business presentations.
• Singing avatar content — Turn a portrait into a lip-synced singing performance.
• Short-form creator videos — Produce avatar content for social media and promotional clips.
• Personalized spokesperson media — Generate custom speaking videos without filming.

Pro Tips

• Use a clear, front-facing portrait for better facial stability.
• Upload clean audio for stronger lip sync and more natural rhythm.
• Keep prompts simple and focused on expression or motion style.
• Use 720p when realism matters more, and 480p when you want lower cost.
• Reuse the same seed when comparing prompt variations on the same portrait.

Notes

• Maximum clip length per job is 64 seconds
• Audio longer than 64 seconds is automatically trimmed
• Generation time is typically around 10–30 seconds of wall time per 1 second of video, depending on resolution and queue load
• Better source portraits and cleaner audio generally improve overall quality

Related Models

• InfiniteTalk — Higher-end avatar video workflow for more advanced speaking performance.
• InfiniteTalk Multi — Multi-character avatar workflow for more complex scenes.

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'
{
  "image": "https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg",
  "audio": "https://interactive-examples.mdn.mozilla.net/media/cc0-audio/t-rex-roar.mp3",
  "resolution": "480p"
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/longcat-avatar-1.5" \
  -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