Mixture-of-Depths: Dynamically allocating compute in transformer-based language models Paper โข 2404.02258 โข Published Apr 2 โข 104
Idefics2 ๐ถ Collection Idefics2-8B is a foundation vision-language model. In this collection, you will find the models, datasets and demo related to its creation. โข 11 items โข Updated May 6 โข 89