LTX-2: 19B Joint Audio-Visual Foundation Model by Lightricks

LTX-2 is Lightricks' 19B parameter DiT-based audio-video foundation model for synchronized video and audio generation with open weights.

L

LTX-2

VideoAudio-VideoText-to-VideoImage-to-VideoDiTLightricks

Next-generation DiT-based audio-video foundation model by Lightricks. Generates synchronized video and audio within a single 19B parameter model. Requires Gemma 3 12B text encoder. Comfy-Org repackaging includes LoRAs and quantized text encoder variants.

Hugging Face (Comfy-Org)Upstream (Lightricks)GitHubPaper
DeveloperLightricks
Release Date2025-12
ArchitectureDiT (19B parameters)
LicenseApache-2.0
CapabilitiesText-to-video, image-to-video, video-to-video, audio-to-video, text-to-audio
Required Text EncoderGemma 3 12B IT

Guides and workflows related to this model series.

No articles found.