CODEX / SIGNAL STUDIO
Google Gemma says EmbeddingGemma 2 delivers best-in-size-class results on benchmarks such as MTEB (Code) and MIEB (Lite), with a 9.92-point code-performance improvement over EmbeddingGemma.
Google detailed operational specifications for EmbeddingGemma 2, stating the 740M model uses between ~191MB and 567MB of active RAM, expands its context window to 8K, and processes up to 5.5 minutes of audio, 29 images, or 58 video frames per pass.
Google for Developers released weights for EmbeddingGemma 2 on Hugging Face and Kaggle, alongside out-of-the-box support for LiteRT, MediaPipe, LangChain, and llama_index.
