LTX 2.3 Finetuned I2V

A hosted browser demo running a finetuned version of the LTX 2.3 image-to-video model. Drop in a still image, get a short animated clip. The image-to-video category is getting crowded fast — Wan, CogVideoX, and the base LTX models are all reasonable comparisons. What is impossible to evaluate here is what the finetune actually optimizes for: there is no readme, no training details, and no benchmark comparison. That opacity is the main reservation. If you are surveying the open-weight I2V landscape before committing to a pipeline, this is worth a five-minute test, but treat it as an anonymous model variant, not a vetted release. Do not build a production workflow around something with no provenance documentation. -> Best for: AI engineer doing a quick capability sweep of open-weight image-to-video models