LLaVA

An open vision-language model family built around LLaMA-style backbones for multimodal image-and-text tasks.

Ecosystem

Backfilled from lab-note mentions and entity refresh.