Lightricks, the creator of Facetune and LTX Studio, has launched LTX-2, the world’s first open source AI video foundation model.
Lightricks, the company behind Facetune and LTX Studio, has unveiled LTX-2, the first complete open-source AI video foundation model, marking a significant leap in creative AI. Built for real-world production, the model unites native 4K fidelity, 50 fps motion, and synchronised audio within a single, production-ready system that runs on consumer-grade GPUs.
The company has released core components of LTX-2, including datasets and inference tooling—on GitHub, with model weights set to follow later this fall. This open-source release allows researchers, developers, and studios to customise, extend, and innovate freely, supporting a collaborative ecosystem. API access is rolling out to early partners through the LTX-2 website, with integrations available via Fal, Replicate, and ComfyUI.
“Diffusion models are reaching a point where they no longer just simulate production — they are production. LTX-2 represents that shift: the most complete creative engine we’ve built,” said Zeev Farbman, co-founder and CEO of Lightricks. “It unites synchronised audio and video, 4K fidelity, long-form capability, and radical efficiency in one open, production-ready system, built to empower everyone, from independent creators to enterprise teams.”
Key advances include synchronised audio-video generation, cinematic 4K quality, up to 50% lower compute costs, and creative precision tools such as multi-keyframe conditioning and 3D camera logic.
With LTX-2, Lightricks is democratising 4K AI video creation, bridging accessibility and innovation through open source collaboration, redefining how creative AI systems are built, shared, and scaled.
