LTX-2.5 text-to-video model released on HuggingFace

On HuggingFace, the model yuvraj108c/LTX-2.5 has been published with the text-to-video task. The publication date is August 12, 2026. As of this news note, the model has 1,445 downloads and zero likes. The code has been released as open source.

What it means

The CPU3D reference currently has no tools with the text-to-video task. The closest entries in terms of topic are 3D generators: OpenLRM, Hunyuan3D 2.1, and TRELLIS. All of them solve the image-to-3d task, meaning they take images as input and produce 3D models as output, rather than working with a text description and video. There is no direct overlap with LTX-2.5: none of the fact sheets describe video generation. So the appearance of this model does not affect the current tool pages in the reference. If text-to-video becomes a separate direction, LTX-2.5 may be considered as a candidate for addition, but for now there is no corresponding category in the section. The model was published recently, and the open description does not yet provide enough data about its capabilities, hardware requirements, and generation quality for an assessment.