HighCapability
Google releases EmbeddingGemma, a new efficient multilingual embedding model
Action Required
Organizations need to migrate to EmbeddingGemma to leverage its high-quality multilingual embeddings for applications like semantic search and RAG pipelines.
AI Impact Summary
Google has released EmbeddingGemma, a new efficient multilingual embedding model with 308M parameters and a 2K context window. Trained on a massive dataset of 320 billion tokens across 100+ languages, it achieves state-of-the-art performance on the Massive Multilingual Text Embedding Benchmark (MMTEB) while maintaining a small memory footprint. This model is designed for on-device use cases like RAG pipelines and agents, offering a practical and open-source solution for multilingual embedding tasks.
Google releases EmbeddingGemma, a new efficient multilingual embedding model
Affected Systems
EmbeddingGemma
- Date
- Date not specified
- Change type
- capability
- Severity
- high