OpenAI's most advanced embedding model with 3072 dimensions, achieving 54.9% on MIRACL benchmark with Matryoshka learning for flexible dimension reduction.
Top open-source multilingual embedding model by BAAI, supporting 100+ languages, 8192 token input length, with unified dense, multi-vector, and sparse retrieval capabilities.
Voyage AI's latest SOTA general-purpose embedding model, ranking first across 8 evaluation domains spanning 100 datasets, outperforming OpenAI and Cohere by 9.74% and 20.71% on average.
