01
97.0
Cohere Embed v3
CohereState-of-the-art embeddings for enterprise search and RAG.
Exceptional retrieval performanceMultilingual supportCustomizable embedding dimensions+1 more
Try Cohere Embed v3Today's top 10 best AI models for Embeddings & Search
02
93.0
OpenAI Ada-002
OpenAIWidely adopted, general-purpose embeddings for diverse applications.
Ease of useBroad compatibilityCost-effective for many use cases+1 more
Try OpenAI Ada-00203
92.0
E5-large-v2
Microsoft ResearchRobust open-source embeddings for search and RAG.
Strong performance on retrieval tasksOpen-source accessibilityGood for multilingual data+1 more
Try E5-large-v204
91.0
Voyager-2
AI2 (Allen Institute for AI)Advanced embeddings focused on factual retrieval and knowledge.
Excellent for knowledge-intensive tasksHigh accuracy in retrievalOpen-source+1 more
Try Voyager-205
90.0
BGE-large-en-v1.5
BAAI (Beijing Academy of Artificial Intelligence)High-performance embeddings for English text.
Top-tier performance on English retrievaOpen-sourceCompetitive with leading proprietary mod+1 more
Try BGE-large-en-v1.506
88.0
Sentence Transformers (all-MiniLM-L6-v2)
UKP Lab (University of Darmstadt)Efficient and effective sentence embeddings for general tasks.
Excellent speed-performance ratioSmall model sizeEasy to use and integrate+1 more
Try Sentence Transformers (all-MiniLM-L6-v2)
07
87.0
Google Gecko AI Embeddings
GoogleEfficient embeddings for on-device and edge applications.
Optimized for mobile and edge devicesLow latencyGood energy efficiency+1 more
Try Google Gecko AI Embeddings08
86.0
Instructor Embeddings
NbAiLab (National University of Singapore)Instruction-tuned embeddings for precise task control.
High task-specific performanceFine-grained control via instructionsOpen-source+1 more
Try Instructor Embeddings09
84.0
Amazon Titan Embeddings G1 - Text
Amazon Web Services (AWS)Enterprise-grade embeddings integrated with AWS.
Seamless AWS integrationScalability and reliabilityGood for general search and RAG+1 more
Try Amazon Titan Embeddings G1 - Text10
83.0
GTE-large
Helsinki-NLP / yoinkEfficient and multilingual embeddings from open-source efforts.
Strong multilingual capabilitiesGood performance for its sizeOpen-source+1 more
Try GTE-large