awesome-foundation-and-multimodal-models

Why we included this project

This index collects notable foundation and multimodal models, with each entry linking to the original paper, the code, and often a live demo or tutorial. For anyone working in vision-language AI, it saves the chore of hunting through arXiv and GitHub separately: models like YOLO-World and Depth Anything appear in one place with their modalities and the tasks they handle. Entries pair each model with runnable resources, so you can quickly judge whether an approach is worth trying for your own detection, segmentation, or captioning work. It is a reference and learning resource rather than software you deploy, which is what makes it useful for evaluation and research.