daVinci-MagiHuman is an advanced, open-source 15B-parameter AI model developed by Sand.ai and GAIR Lab at Shanghai Jiao Tong University. It is designed to generate high-quality, lip-synced talking videos from a single portrait image and a script or audio file. Unlike traditional methods that combine separate text-to-speech and video pipelines, daVinci-MagiHuman utilizes a unified single-stream Transformer to jointly denoise video and audio tokens simultaneously. Released under the Apache 2.0 license, it allows users to inspect weights, run inference locally, and use the technology for commercial purposes. It is optimized for speed, capable of generating short clips in just seconds on professional-grade hardware like the NVIDIA H100.
Open-source AI generating lip-synced talking videos from a single photo and audio/text.
- Creating AI-powered marketing avatars from static portraits
- Developing multilingual educational content with synchronized lip motion
- Generating low-latency digital humans for interactive applications
- Prototyping realistic talking head animations for social media
- To use daVinci-MagiHuman
- upload a clear
- front-facing portrait photo and provide a script or audio file. Select your desired output resolution (e.g.
- 256p
- 720p
- or 1080p) and start the generation process. Once the AI completes the job
- you can download your talking video. For local deployment
- users can download the model checkpoints from Hugging Face and follow the provided CLI instructions.
AI tool for realistic talking avatars and speech animations.
Create realistic AI avatars and talking videos from images.
No-code AI platform for creating AI workflows and generating videos with AI avatars.
AI platform for creating personalized videos at scale.
Video platform for sales and marketing to increase leads and customer engagement.
Avatar tool for micro YouTubers to create engaging video content easily.
AI-powered tools for game development and content creation, including portrait animation.
AI studio for creating viral, avatar-led short-form videos from scripts.
Hyper-realistic AI Video Agents for real-time, human-like interactions and automated conversations.
