v4.56.0-Embedding-Gemma-preview
📦 transformers
✨ 3 features🔧 2 symbols
Summary
This release introduces a preview of the EmbeddingGemma model, a highly efficient 308M parameter multilingual embedding model optimized for on-device RAG and mobile use cases.
Migration Steps
- Install the specific preview tag using: pip install git+https://github.com/huggingface/transformers@v4.56.0-Embedding-Gemma-preview
✨ New Features
- Added support for EmbeddingGemma, a state-of-the-art multilingual embedding model.
- Support for 100+ languages with a compact 308M parameter size and 2K context window.
- Integration with sentence-transformers for query and document encoding.
🔧 Affected Symbols
EmbeddingGemmaSentenceTransformer