MiniMax-H3-slim released on HuggingFace for text-to-video generation

On August 24, 2026, the model fkyyy/MiniMax-H3-slim was published on HuggingFace with the text-to-video task. At the time of publication, it has 508 downloads and no likes. Details are on the model page.

What it means

The model belongs to the text-to-video task, whereas our section AI for 3D and video on 3D model generation describes tools with the image-to-3d task: OpenLRM, Hunyuan3D 2.1 and TRELLIS. None of them solves the text-to-video task, so MiniMax-H3-slim cannot be compared with them directly. According to the HuggingFace page, the developer specifies neither VRAM requirements nor export formats. The code is released as open source, but the launch parameters and limitations of the model are not disclosed in the description. For those looking for 3D model generation tools, MiniMax-H3-slim is not a replacement: it generates video, not three-dimensional assets. If what you need is specifically text-to-video, it is worth waiting for more detailed documentation from the model's author.