Hey everyone! I just launched ModelRegistry on Product Hunt today.
It’s an open index designed to track foundation model flagships, context limits, and research checkpoints across major AI labs in one centralized place.
@Google dropped EmbeddingGemma 2 (Apache 2.0).
Maps text, code, images, video, and audio into a single 768-dim vector space on-device in under 600MB RAM.
Local cross-modal RAG without running 3 separate embedders is finally clean.
