mmastrac runs EmbeddingGemma2 as a live /v1/systemone endpoint: ~20ms text, ~100ms images, distillable and multimodal.
Matt Mastracci @mmastrac
EN
- 47.5k views
- 876 likes
- 772 bookmarks
- 56 reposts
About this case
Platform note with media. mmastrac shows Google DeepMind EmbeddingGemma2 wired as an insanely fast real-time /v1/systemone endpoint (20ms text, 100ms images), trainable in real time, able to distill an upstream teacher, multimodal by default.
Original posthttps://x.com/mmastrac/status/2115098609356284223