Live-portrait

Live-portrait

Live Portrait is an AI-driven portrait animation framework that transforms a single static photo into lifelike animated videos, producing realistic facial expressions, smooth head movement, and precise lip synchronization with fine-grained control.

Live-portrait is video generation software teams evaluate for video generation. Use this page to review pricing, integration signals, and the best alternatives before you commit.

Free API 70/100
#171 in Video Generation (171 tools)
Added 5 months ago
Data reviewed Jul 16, 2026

Profile facts come from the vendor source. AiMatch labels unknown pricing or API details instead of estimating them.

Review official source →

Quick Overview

Best for: Video Generation

What it does

Video Generation software for decision-makers comparing workflow fit and alternatives.

Best fit

Video Generation

Pricing snapshot

Free

Next step

Compare Live-portrait with similar tools before you shortlist it.

Compare this tool before you shortlist it

Review alternatives, pricing posture, and workflow fit side by side.

Live-portrait

Live Portrait is an AI-powered portrait animation framework that converts a single static portrait image into a realistic animated video. It produces controllable facial expressions, accurate lip movements, and smooth head motion using an implicit keypoint animation architecture and stitching/retargeting modules. The project emphasizes efficiency and practical usage — offering a free playground for experimentation and inference code and models on GitHub — making it suitable for researchers, developers, and creators who need production-oriented portrait animation and editing tools.

Live Portrait is an AI-driven portrait animation framework that transforms a single static photo into lifelike animated videos, producing realistic facial expressions, smooth head movement, and precise lip synchronization with fine-grained control.

Own this listing?

Claim this page to add pricing, features, screenshots, and verified owner details.

Claim this listing

Key Features

Multi-Style Portrait Animation

Animates still images across various styles and sizes using stitching technology; supports realistic and stylized outputs (e.g., oil painting, sculpture, 3D rendering).

Portrait Video Editing with Stitching

Provides dynamic video-to-video face animation and editing capabilities using stitching modules to refine outputs.

Precise Eye Animation Control

Dynamically adjusts eye openness using scalar inputs for nuanced control over eye expressions.

Accurate Lip Movement Control

Fine-tunes lip positions with scalar-based inputs to achieve precise speech and expression animation.

Advanced Self-reenactment

Generates high-quality animations from a single source frame, designed to outperform existing methods in recreating dynamic video sequences.

Powerful Cross-reenactment

Transfers motion between different portraits robustly, handling large pose changes, subtle expressions, and multi-person inputs.

Implicit Keypoint Animation

Maps motion from a driving video onto the portrait using an implicit keypoint framework for controllable animation.

High Efficiency Video Synthesis

Optimized decoder rapidly generates frames to create smooth animations with high speed (reported 12.8ms per frame on an RTX 4090 using PyTorch).

Stitching and Retargeting Modules

Small MLP-based modules for refining outputs and providing fine-grained control and retargeting with minimal overhead.

Pricing

Free Tier Available

Free playground available at https://live-portrait.org/free-playground ("Start for Free" is shown on the site).

Use Cases

Self-reenactment

Animate a subject using motion derived from the same person’s driving video to recreate expressions and pose dynamics.

Cross-reenactment

Transfer motion from one person (driving video) to a different portrait to animate a target subject.

Portrait Animation from Still Images

Synthesize lifelike motion and expressions from a single static portrait photo.

Portrait Video Editing

Edit and retarget facial motion within existing portrait videos using stitching and retargeting modules.

Eye and Lip Retargeting Control

Precisely control eye openness and lip positions for nuanced expression or speech animation.

Style Transfer and Multi-style Animation

Render animations in different visual styles (realistic, oil painting, sculpture, 3D renderings).

Animal Animation (fine-tuning)

Can be fine-tuned to animate animals (cats, dogs, pandas) as noted in application scenarios.

Integrations

PyTorch

Framework used for model inference and performance reporting (implementation and timings reported using PyTorch).

GitHub (KwaiVGI/LivePortrait)

Inference code and pretrained models are hosted on GitHub for download, reproduction, and integration (https://github.com/KwaiVGI/LivePortrait).

NVIDIA GPUs (RTX 4090)

Performance benchmark reported on RTX 4090 hardware (12.8ms/frame) indicating GPU-accelerated inference compatibility.

Benefits

Generate lifelike animated videos from a single static photo, enabling realistic facial expressions and head motion.
Fine-grained control over eye, lip, and pose retargeting for precise expression editing.
High efficiency and speed for practical usage (reported 12.8ms/frame on RTX 4090) and availability of open-source inference code for integration and customization.

Limitations

Claim this listing to add transparent limitations.

Frequently Asked Questions

What is Live Portrait?
Live Portrait is a video-driven portrait animation framework that focuses on better generalization, controllability, and efficiency for practical usage. It's designed to synthesize lifelike videos from a single source image.
What are the main features of Live Portrait?
Main features include generating realistic videos from a single source image with motion derived from driving videos, audio, text, or generation; supports self-reenactment, cross-reenactment, eye and lip retargeting control, and multiple visual styles.
Who developed Live Portrait?
Live Portrait was developed by researchers from Kuaishou Technology, University of Science and Technology of China, and Fudan University.
What core technologies does Live Portrait use?
It uses an implicit-keypoint-based framework (built on vid2vid), mixed image-video training strategy, upgraded network architecture, motion transformation and optimization objectives, and stitching/retargeting modules with small MLPs.
What are the application scenarios for Live Portrait?
Application scenarios include self-reenactment, cross-reenactment, portrait animation from still images, portrait video editing, eye and lip retargeting control, animation across different styles, and fine-tuning for animal animation.
What is the performance of Live Portrait?
Reported generation speed is 12.8ms on an RTX 4090 GPU using PyTorch. The project used about 69 million high-quality frames for training to improve quality and generalization.
How can I get more information about the Live Portrait open-source project?
Inference code and models are available at https://github.com/KwaiVGI/LivePortrait; more detailed information and experimental results are on the project's GitHub page.
How can I start using Live Portrait for free?
To start using Live Portrait for free, visit the free playground at https://live-portrait.org/free-playground.

Getting Started

  1. 1 Step 1: Upload a clear portrait image — Live Portrait extracts facial features from the source photo.
  2. 2 Step 2: Select or upload a driving video that contains the facial movements to apply to the portrait.
  3. 3 Step 3: Click 'Animate', optionally adjust configuration parameters, then generate and wait for the animated video to be produced.

Support

email

Contact via the site email: [email protected] (listed in site footer).

docs

Inference code, models, and project details are available on the GitHub repository: https://github.com/KwaiVGI/LivePortrait.

web-playground

Try the tool directly via the Free Playground at https://live-portrait.org/free-playground.

API

Available: No
Documentation:

https://github.com/KwaiVGI/LivePortrait (inference code and models hosted on GitHub; no API described on the site)

Compare Live-portrait with similar tools

See how it stacks up against alternatives

Related Tools

View all 171 →
Freemium
Aianimateimage

Aianimateimage

Animate Image AI transforms static photos into short, shareable videos using deep learning models — aimed at creators, marketers, and brands who want to boost social engagement without manual editing.

Video Generation
High-growth
Freemium
Fliki

Fliki

Fliki is an AI-powered text-to-video platform that converts scripts, blog posts, slides, or prompts into finished videos with lifelike AI voiceovers, captions, visuals, and localization across 80+ languages.

Video Generation
Freemium
image-to-video-ai-imagemover

image-to-video-ai-imagemover

ImageMover is a web-based AI Image-to-Video generator that converts static photos (PNG, JPG, WEBP) into MP4 videos using multiple AI models and customizable settings, aimed at creators, marketers, and businesses.

Video Generation
High-growth
Freemium
tooncrafter

tooncrafter

ToonCrafter is an AI-powered animation tool that turns static cartoon images or keyframes into smooth, stylized animated MP4 clips, aimed at creators, marketers, and designers who want quick, production-ready cartoon animations.

Video Generation
High-growth
Free
Viroo

Viroo

Viroo.ai is an AI video generator that transforms text, images, or simple inputs into short, social-first videos (TikTok, YouTube) with cinematic effects — aimed at creators, e-commerce advertisers, and agencies.

Video Generation
Contact for pricing
Clipfly

Clipfly

Clipfly is an all-in-one AI-powered creative platform offering online video editing, AI video/image generation, music generation, and a wide set of editing tools for producing videos, images, and audio.

Video Generation
Contact for pricing
Fast-wan

Fast-wan

Fast Wan is an AI video generation platform offering professional-grade video creation using its Wan 2.2 and Wan 2.1 models, targeting creators and teams who need high-quality, fast AI-generated videos.

Video Generation
Contact for pricing
Aiimageanimator

Aiimageanimator

AI Image Animator is a web-based tool that turns still images into short animated videos by uploading an image, entering a prompt, and generating a moving MP4 clip using AI-driven animation models.

Video Generation

Premium Alternatives

Paid
Shuffll

Shuffll

Shuffll is an enterprise-focused video infrastructure platform that automates generation of thousands of on‑brand videos from structured data via API, enforcing brand governance and embedding video capabilities directly into platforms and workflows.

Video Generation
Enterprise-ready
Paid
alle-ai

alle-ai

Alle-AI is an all-in-one generative AI platform that combines and compares outputs from multiple AI models to produce more accurate, trustworthy results and offers multimodal generation (images, audio, video) for individuals, businesses, education, and developers.

Video Generation
Enterprise-ready High-growth
Paid
Veo4aivideo

Veo4aivideo

Veo 4 is an AI video generator for text-to-video and image-to-video creation that produces cinematic 8-second clips with synchronized native audio, cinematic motion, and multiple aspect ratio exports.

Video Generation
High-growth
Paid
shorts-faceless

shorts-faceless

ShortsFaceless is an AI-powered platform that automates creation of faceless short-form videos (YouTube Shorts, TikTok, Reels) by generating scripts, images, voiceovers, subtitles and exporting HD videos to help creators scale production quickly.

Video Generation
High-growth
Paid
Dreaminai

Dreaminai

Dreamina AI is an online AI-powered video and image generator that converts text prompts and reference images into production-ready cinematic videos and images, aimed at marketers, creators, and teams. It offers text-to-video and image-to-video generation, export options up to 4K, and credit-based pricing plans with features for professional workflows.

Video Generation
Enterprise-ready
Paid
Kling3

Kling3

Kling 3 is an AI-powered video and image generation platform (Kuaishou's third-generation model) that creates cinematic 4K images and up to 15-second videos with character consistency, multilingual lip-sync, and integrated audio.

Video Generation
Paid
Veo3-2

Veo3-2

Veo 3.2 is an AI-powered video generation model that converts reference images into expressive, high-quality videos with features like character and object consistency, native vertical (9:16) output, and 4K upscaling.

Video Generation
Paid
Kling3

Kling3

Kling 3 AI is a web-based text-and-image to cinematic video generator that uses advanced neural networks to produce ultra-HD, studio-quality videos with realistic motion, camera control, and scene composition for marketers, creators, and businesses.

Video Generation
Enterprise-ready

Explore Related Categories

Explore by Outcome