§ models · storyline
Google DeepMind launches EmbeddingGemma 2 for multimodal embeddings
Google DeepMind launches EmbeddingGemma 2, a 740M-parameter on-device model for mapping text, images, video, and audio to shared embeddings under Apache 2.0 license.
Google: Google DeepMind launches EmbeddingGemma 2, a 740M-parameter model to map code, images, video, and audio in a shared embedding space, under an Apache 2.0 license — EmbeddingGemma 2 is the most capable model for on-device multimodal embeddings, natively mapping combinations of text, images …
§ sources4 publications · timeline below
- blog.googleGoogle DeepMind launches EmbeddingGemma 2, a 740M-parameter model to map code, images, video, and audio in a shared embedding space, under an Apache 2.0 license (Google)primary
- deepmind.googleEmbeddingGemma 2: an open, lightweight multimodal embedding model
- the-decoder.comGoogle claims EmbeddingGemma 2 outperforms rival embedding models twice its size
- github.comTransformers: Release v5.19.0
§ how this story moved
- primary — Blog publishes the launch post.
- transformers — Releases picks up coverage.
- The Decoder picks up coverage.
- Google DeepMind — Blog picks up coverage.