On March 5, 2025, the Lightricks team officially released the LTX-Video version 0.9.5, an open source AI video generation model based on DiT (Diffusion Transformer) technology, marking a major breakthrough in the field of AI video creation. As the latest iteration of the world's first real-time high-quality video generation model, LTX-Video 0.9.5 not only achieves a significant improvement in technology, but also provides more possibilities for developers and enterprise users through open commercial licenses.
Lightricks officials announced the news through social media that morning and detailed the core update of the new version. LTX-Video 0.9.5 not only has significantly improved the generation quality, but also added a number of practical functions, further promoting the development of open source AI video creation. The release of this version undoubtedly brings a brand new creative experience to technology enthusiasts and business users around the world.
One of the biggest highlights of LTX-Video 0.9.5 is its support for commercial licensing, which means that enterprises and individual developers can apply it to commercial projects, greatly expanding their application scenarios. In addition, the new version also introduces keyframe condition support functions, where users can directly generate content from the first, last or intermediate frames of the video, providing greater flexibility for video expansion and editing. The addition of this function makes video creation more efficient and convenient.
In terms of resolution and generation quality, LTX-Video 0.9.5 also achieved significant improvements. The new version supports up to 768x512 resolution and real-time generation of 24 frames per second, while reducing artifacts and making the picture smoother and more natural. In addition, the model also enhances the generation ability of long video sequences, which can better meet the needs of complex narratives. Native integration with ComfyUI allows users to seamlessly integrate into existing workflows, further improving operational ease.
LTX-Video 0.9.5 has been opened for download on GitHub and Hugging Face platforms, and supports Python 3.10.5, CUDA12.2 and other environments. The community generally believes that this update further consolidates LTX-Video's leading position in the field of AI video generation. In the future, with further optimization of generation speed and resolution, LTX-Video is expected to realize its potential in more scenarios. As Lightricks said, LTX-Video 0.9.5 brings "open source AI video creation to new heights", opening a door to the future for global creators.