GPT Image 1.5 Edit is now live!
🎥 Video Generation

Live Avatar

Real-time avatar generation with natural face-to-face conversations. Stream infinite-length video with immediate visual feedback, synchronized to audio input

Example Output

Input

Input Example
Original

Output

Generated

Instructions

"A person speaking naturally with expressive gestures."

Try Live Avatar

Fill in the parameters below and click "Generate" to try this model

Reference image for avatar (character will be animated)

Driving audio file (WAV or MP3, avatar syncs to this)

Scene and character description

Acceleration level for faster decoding

Your inputs will be saved and ready after sign in

More Video Generation Models

Google Veo 3.1 Reference-to-Video

Create videos using multiple reference images for consistent subject appearance.

DoP Image-to-Video

DoP Image-to-Video

Animate static images into 5-second videos with zoom, pan, and rotate effects.

Kling 1.6 Pro Image-to-Video

Animate images with superior motion quality and ending frame control

Sana Video

Create videos from text at lightning speed with motion control

Seedance 1 Lite

Generate videos from text or images up to 10s long in 720p

Google Veo 3.1 Image-to-Video

Animate images into high-quality videos with sound.

Stable Video Diffusion

Stable Video Diffusion

Turn images into smooth videos with adjustable motion and frame rate controls

Kling 1.6 Pro Text-to-Video

Turn text into videos with enhanced quality and fine details

Leonardo Motion 2.0

Turn text into 5s videos with style controls and smooth frame interpolation