{"schema":"postcutoff/model@1","as_of":"2026-11-19T22:11:00+01:00","url":"https://postcutoff.com/m/ltx-2-5/","md":"https://postcutoff.com/m/ltx-2-5/index.md","disclosure":{"written_by":"AI agents (Claude Opus 5.5 in Claude Code)","editor":"human editor","policy":"https://postcutoff.com/about/"},"license":null,"id":"ltx-2-5","name":"LTX-2.5","org":"Lightricks","family":"LTX-2","released":"2026-08-11","status":"current","type":"video-gen","modality_in":["text","image","video"],"modality_out":["video","audio"],"open_weights":true,"model_license":"ltx-2-community-license (free under $10M annual revenue)","context_window":null,"max_output":null,"knowledge_cutoff":null,"pricing":{"fast_per_second":0.09,"4k_per_second":0.3,"unit":"USD per second of output video on Lightricks' managed API (as reported by VentureBeat)","source":"https://venturebeat.com/technology/ltx-2-5-can-generate-a-10-second-ai-video-from-an-image-in-just-6-8-seconds-on-nvidia-superchips-and-its-open-weights"},"price_line":"$0.09 per second, fast","access":[{"provider":"Hugging Face","url":"https://huggingface.co/Lightricks/LTX-2.5"},{"provider":"GitHub","url":"https://github.com/Lightricks/LTX-2"},{"provider":"LTX API","url":"https://console.ltx.io/playground/","docs":"https://docs.ltx.io"}],"capabilities":[{"name":"Open audio-video generation with multishot consistency","detail":"22B DiT that generates synchronized video and audio and keeps character, environment, lighting and voice consistent across cuts.","first":false,"discovered":"launch","source":"https://huggingface.co/Lightricks/LTX-2.5"},{"name":"Fast generation","detail":"10-second 720p image-to-video clip in 6.8 s on two GB200s (vendor claim); distilled variant uses 8 steps.","first":false,"discovered":"launch","source":"https://venturebeat.com/technology/ltx-2-5-can-generate-a-10-second-ai-video-from-an-image-in-just-6-8-seconds-on-nvidia-superchips-and-its-open-weights"},{"name":"Robotics / physical-AI checkpoint","detail":"Ships a pretrained checkpoint aimed at world-model use in robotics and physical AI.","first":false,"discovered":"launch","source":"https://venturebeat.com/technology/ltx-2-5-can-generate-a-10-second-ai-video-from-an-image-in-just-6-8-seconds-on-nvidia-superchips-and-its-open-weights"}],"entry":"2026-08-11-lightricks-ltx-2-5-open-weights","notes":"Full (trainable) and distilled DiT variants; Gemma 4 12B text encoder; up to 4K; frame counts must satisfy n % 8 == 1; ComfyUI native. Licence bans military/weapons use.","verified":"2026-10-03","body_md":"Lightricks' open-weights video+audio model; the most downloaded generative model on Hugging Face in early October 2026.","page_url":"https://postcutoff.com/m/ltx-2-5/","standing":{"kind":"video generation","newer_count":7,"newer":[{"id":"gemini-omni-1-2-flash","name":"Gemini Omni 1.2 Flash","org":"Google DeepMind","released":"2026-11-05","page_url":"https://postcutoff.com/m/gemini-omni-1-2-flash/"},{"id":"flux-3-max","name":"FLUX 3 Max","org":"Black Forest Labs","released":"2026-10-19","page_url":"https://postcutoff.com/m/flux-3-max/"},{"id":"kling-4-0","name":"Kling 4.0 (and 4.0 Flash)","org":"Kuaishou","released":"2026-10-15","page_url":"https://postcutoff.com/m/kling-4-0/"},{"id":"tavus-griffin","name":"Tavus Griffin (Human Interaction Model)","org":"Tavus","released":"2026-10-01","page_url":"https://postcutoff.com/m/tavus-griffin/"},{"id":"heygen-video-1","name":"HeyGen Video 1","org":"HeyGen","released":"2026-09-30","page_url":"https://postcutoff.com/m/heygen-video-1/"}],"before":null,"after":null,"weights":"public","licence":"ltx-2-community-license (free under $10M annual revenue)","licence_permissive":false,"summary":"Weights are public. Licence: ltx-2-community-license. See the licence for commercial terms."}}