Gemini Embedding
Google's unified embedding model — topped the MTEB multilingual leaderboard at launch
Verdict
Google consolidated its embedding lineup into a single Gemini-based model that debuted at the top of the MTEB multilingual leaderboard. Matryoshka representation learning support, strong cross-lingual retrieval. Worth a serious look alongside text-embedding-3-large for new RAG builds.
Other Embedding Models
- BGE-M3Stable
BAAI multi-granularity multilingual embedding with dense + sparse
- Cohere Embed v3Production
Multilingual embedding model optimised for search and RAG
- E5-Mistral-7BStable
Large LLM-based embedding model for maximum retrieval quality
- GTE-Qwen2Experimental
Alibaba's embedding model with strong CJK language support
- Jina Embeddings v3Stable
8K context multilingual embedding with task-specific LoRAs
- Mixedbread EmbedExperimental
Emerging high-quality embedding model from Berlin-based lab