LTX-2
Lightricks' open-weight video model with synchronized sound
Lightricks' open-weight video model with synchronized sound
LTX-2 is an AI video model from Lightricks, the company behind the LTX Studio storytelling platform. Lightricks has released LTX-2 with open weights, so technical teams can download it and run it on their own hardware, including consumer GPUs. It is also available through hosted services.
LTX-2 is known for generating video and synchronized audio in a single pass, so effects, ambience and speech can line up with the picture. Lightricks also highlights high resolution output and fast rendering, with a distilled version for quicker drafts. That makes it interesting for self-hosted pipelines and for drafts where timing to sound matters.
LTX-2 is not available in Cinemagiq. For video with sound, MiniMax H3 Max is fast, follows prompts tightly and always generates sound, up to 1080p. Grok Imagine Video 1.5 works from a prompt, a first frame or up to 7 references, with native audio on every clip. PixVerse C1 is a fast option for drafts, with optional audio.
LTX-2 isn't available in Cinemagiq. These models cover similar work.
No. LTX-2 is not available in Cinemagiq. The Generation Studio offers other models that generate sound with the picture, such as MiniMax H3 Max and Grok Imagine Video 1.5, plus PixVerse C1 with optional audio for fast drafts.
MiniMax H3 Max is fast, follows prompts tightly and always adds sound, from 480p to 1080p. Grok Imagine Video 1.5 runs 1 to 15 seconds with native audio from a prompt, a first frame or references. PixVerse C1 suits quick drafts with optional audio.
LTX-2 is known for generating video and synchronized sound together and for fast rendering. Because Lightricks released open weights, it suits teams that run their own video pipelines, want to customize a model, or need drafts timed to sound.
LTX-2 is made by Lightricks, the company behind Facetune and LTX Studio. It is available as open weights and through hosted services. It is a separate tool, so clips made with it are not part of your Cinemagiq project.
Cinemagiq runs the leading video, image and audio models in one AI-native workspace, so you can pick the right one per shot.