V

Vidu S1 API

Streaming video generation model and API for real-time interactive digital humans

freemium
Last Updated 9/11/2026

About Vidu S1 API

Vidu S1 is a streaming video generation model for real-time interactive digital humans. It generates video during conversations: users speak, the character can see and hear them, and responds with expressions, voice, and personality in near real time. Integration can use HTTP to manage sessions, AliRTC to transmit audio and video, and WebSocket to handle control signaling; it also supports single-person images and persona prompt to define the character's identity. The voice library offers 50+ voices, each supporting 28 languages, and can clone your own voice.

Best for

Real-time interactive digital human developers
Streaming video application developers
Voice character product teams

Best use cases

Real-time digital human conversations

Interactive character experiences

Multilingual voice characters

Video generation API integration

Features

Real-time streaming video

Generate live video during conversations with expressive, vocal, and personalized character presentation.

Three-channel integration

Use HTTP to manage sessions, AliRTC to transmit audio and video, and WebSocket to handle control signaling.

Character image and setup

Support single-person full-body or half-body PNG, JPG, JPEG, or WEBP images, combined with persona prompt to define the character's appearance and behavior.

Multilingual voice

Offer 50+ voices, each supporting 28 languages, with the option to clone your own voice.

Pros

  • Generate video in real time during conversations
  • Support expressive, vocal, and personalized character performance
  • 50+ voices covering 28 languages
  • HTTP, AliRTC, and WebSocket integration channels

Cons

  • A single session is limited to 600 seconds and disconnects automatically afterward.
  • The server automatically closes the connection when the balance reaches zero.

Pricing Plans

Free Trial

1,000 creditstrial
  • Provide a trial allowance for each new user
  • Support about 11 minutes of real-time interaction

Pay As You Go

3 credits / 2 susage
  • Billing starts once the character goes live
  • The server automatically closes the connection when the balance reaches zero

* Pricing information is for reference only. Please refer to the latest announcement on Vidu S1 API's official website.

FAQ

A single session lasts up to 600 seconds; the server disconnects automatically afterward. For a longer experience, create a new session and reconnect.
Supports a single-person full-body or half-body image in PNG, JPG, JPEG, or WEBP, up to 50MB, provided via URL or Base64.

Tool Information

DeveloperVidu S1 API
Release Date9/11/2026
Pricingfreemium