The First Live-Stream Diffusion AI Video Model with <40ms Response Time
Transform infinite video streams in real-time. Zero latency, infinite generation, no quality degradation.
Real-time processing at 24 FPS with <40ms response time. No delays, no buffering, just instant transformation.
Using Diffusion Forcing technology, generate infinite video without quality degradation or error accumulation.
Access technical papers, CUDA kernel documentation, and implementation details. Join our research community.
Choose your platform and download the latest version with SHA256 verification.
Introducing MirageLSD: The First Live-Stream Diffusion (LSD) AI Model
— Decart (@DecartAI) July 17, 2025
Input any video stream, from a camera or video chat to a computer screen or game, and transform it into any world you desire, in real-time (<40ms latency).
Here's how it works (w/ demo you can use!): pic.twitter.com/1o5GgSpBZ6
This new AI "MirageLSD" transforms live video in real-time.
— Mark Kretschmann (@mark_k) July 18, 2025
Like video filters on steroids!
What a time to be alive!
pic.twitter.com/F5wnvM84XH
I love when I test a new AI technology that gives me that "okay, this is magic" feeling.
— Igor Pogany (@IgorPogany) July 17, 2025
I haven't had that feeling in a while, but Decart AI just launched MirageLSD and I'm pretty blown away. MirageLSD is a brand new AI model that transforms live video with zero latency.
It's… pic.twitter.com/1i2N0zugsk
$0
$19/mo
$49/mo