Video model API
Kling Avatar V2 API
kwaivgi/kling-avatar-v2 is Kuaishou's advanced AI talking avatar model, designed to transform a single portrait image and an audio recording into a realistic, lip-synced talking video. It generates natural facial expressions, accurate mouth movements, and smooth head motion while preserving the identity of the original character, making it ideal for creating digital presenters, virtual influencers, educational content, and social media videos. Supporting realistic humans, anime characters, cartoons, and even animals, Kling Avatar V2 delivers high-quality avatar animations without requiring ma
- Model id
kling-avatar-v2- Takes
- Text (optional) + Image (required) + Audio (required)
- Makes
- Video
- Endpoint
POST /api/v1/run- Made by
- Kling (Kuaishou)
- Mode
- Async (job polling)
Call Kling Avatar V2
https://payload.rentprompts.com/api/v1. Replace the placeholder with your key.Model kling-avatar-v2
POST /api/v1/runasync · poll /v1/jobs/:id
curl https://payload.rentprompts.com/api/v1/run \
-H "Authorization: Bearer rp_live_YOUR_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "kling-avatar-v2",
"prompt": "A slow drone shot over a misty pine forest"
}'
# Async models answer HTTP 202 with a job. Poll it until status is
# "succeeded" (output.url / output.urls) or "failed" (error):
curl https://payload.rentprompts.com/api/v1/jobs/JOB_ID \
-H "Authorization: Bearer rp_live_YOUR_KEY"Replace rp_live_YOUR_KEY with your key, ideally via the RENTPROMPTS_API_KEY environment variable. Keys are shown only once, so they are never filled in here. Get your API key →
What a typical call costs
- 5 seconds of output30 Joules
- 10 seconds of output60 Joules
Send requests to POST /api/v1/run. Kling Avatar V2 runs asynchronously: the request answers HTTP 202 with a job, and you poll GET /api/v1/jobs/{id} until the result is ready.
Settings
This model has no extra settings. Send the standard fields shown in the example.