EmbeddingGemma 2
Google's small open model that makes text, images, audio, and video searchable on a phone or laptop, for developers building on-device search.

Details
- Maps text, code, images, audio, and video into one shared space, so a single query can match across formats. The first version handled text only.
- 740M parameters on the Gemma 4 architecture. Text-only use needs as little as 270M parameters, with the vision (170M) and audio (300M) parts optional, about 191MB of RAM on a Pixel 11 Pro.
- Handles 8K tokens per input, four times the first version, which is enough for about 5.5 minutes of audio, 29 images, or 58 video frames.
- Flexible embedding sizes cut storage in a local database by up to 6x. Google also calls out a large jump in code search quality over the first version, aimed at indexing local codebases.
- Open weights under the Apache 2.0 license on Hugging Face and Kaggle, with built-in support for LiteRT, MediaPipe, LangChain, and LlamaIndex.
Best for
Building search that finds a video clip, audio recording, or image from a plain-text query, all on the user's own device.
When to use it
Your app needs to search mixed media without sending files to a server, for privacy or offline use.
Who benefits
App developers adding private or offline search and retrieval to mobile and desktop products.
Join the newsletter to get ahead with AI.
I share strategies, new modules, free guides, and tools to get you better at using AI in your work.
Leading a team? See team programs →