Post

On-device embeddings with EmbeddingGemma 2 in the Koog-based Confetti AI agent @googlegemma johnoreilly.dev/posts/confet... No more paying for embedding API calls!

johnoreilly.devOn-device embeddings with EmbeddingGemma 2 in the Koog-based Confetti AI agentIn an earlier article we added semantic search to the Koog-based AI agent in the Confetti Compose Multiplatform sample. That used Gemini both for the LLM driving the agent and for the embeddings behin...

13/10/26, 7:47 pm0 reposts1 like

John O'Reilly

@johnoreilly.dev

I was also able to use this alongside Gemma 4 (on-device LLM) for a fully offline experience but wanted to firstly show here how easily EmbeddingGemma 2 could be used (independent of the LLM).

John O'Reilly

@johnoreilly.dev

Also really like this image that Claude created to use in the article (based on real data that Confetti has for the conference)