VLM blog posts

Vision-language models (VLMs) connect what a model sees to what it can say. These posts put VLMs to work in FiftyOne: auto-labeling, visual grounding, dataset querying, and hands-on evaluations of new models as they release.