Radar / AI media tools / Lightricks LTX-2.5
Lightricks LTX-2.5
Open-weight 22B world model generating synchronized video + audio from text, image, and video inputs, with native multishot scene generation and a pre-trained checkpoint for robotics fine-tuning.
Why it matters
First open 22B video model with native synchronized audio generation and a separate pre-trained checkpoint explicitly aimed at robotics and world-model fine-tuning.
What you could build with it
An indie video studio could self-host LTX-2.5 to generate synchronized video-plus-audio ad variants from product photos, iterating locally before renting GPUs for final renders, cutting stock-footage and voiceover costs per campaign.
Does it hold up?
Usable today for builders: community ComfyUI/RunPod integrations and an MLX Mac port exist; vendor claims 10s of 720p in 6.8s on 2x GB200, though weights are access-gated on Hugging Face.
Built with Lightricks LTX-2.5
- alterpeace/runpod-comfy: LTX-2.5 supportgithub · Third-party RunPod + ComfyUI deployment pack adding LTX-2.5 support: model manifest with 17 files, HF downloader script, all-in-one installer, and example T2V/V2V workflows.
- comfyui-continuity-mac: LTX 2.5 model docsgithub · Mac-focused ComfyUI pack documenting LTX-2.5 slots (transformer, Gemma 4 text encoder, video/audio VAEs, upscalers, IC-LoRAs) — real integration work by a community builder.
- mlx-community/ltx-2.5-mlxhuggingface · Community MLX-format conversion of LTX-2.5 for Apple Silicon, parity-gated per component against the PyTorch reference — end-to-end generation on Mac.
- alterpeace/runpod-comfygithub · 1 ★ · RunPod + ComfyUI deployment pack with LTX-2.5 installer and example workflows
Learn more
First spotted on hf: source.
More AI media tools
OpenMontage: open-source agentic video production systemTurns an AI coding assistant into a full video studio: 12 pipelines and 100+ tools for scripting, footage, narration…media · JEV 0.73HyperFramesHeyGen's open-source framework that turns HTML/CSS/GSAP into deterministic MP4 video via 21 agent skills and a router…media · JEV 0.69Meshy crosses $100M ARR and launches official iOS/Android appsMeshy announced Sep 30 that ARR surpassed $100M (up 100x in under two years, first AI-3D company at that milestone)…media · JEV 0.61Kandinsky 6.0 Video: open MIT-licensed audio-video generation modelsSber's team released Kandinsky 6.0 Video: Lite (3B) and Pro (29B) diffusion models generating 5-second clips with…media · JEV 0.59FLUX 3 ImageBlack Forest Labs' new multimodal image model with bounding-box layout control, up to 10 reference images, and native…media · JEV 0.57DuoMatching: few-step video generation via joint-marginal distribution matching (ByteDance)Distillation framework pairing joint distribution matching with frame-level marginal supervision from an image…media · JEV 0.51
Get the week's best AI launches, plus 3 ideas worth building
One email every Saturday. Ranked by traction, not hype. Free.