MTEB: Massive Text Embedding Benchmark
Sep 7, 2026
Summary
Introduces the Massive Text Embedding Benchmark (MTEB) for evaluating text embedding models. MTEB provides a standardized framework with over 100 datasets and 50 tasks to assess the performance of embedding models across different domains and use cases, including sentence similarity, retrieval, and classification.
More from the Sentence Transformers for Playlist Generation trail
- NeoMME: an efficient Multimodal-native and Multilingual Encoder
- Introducing OlmoEarth embeddings: Custom embedding exports from OlmoEarth Studio for downstream analysis
- Build a Domain-Specific Embedding Model in Under a Day
- Granite Embedding Multilingual R2: Open Apache 2.0 Multilingual Embeddings with 32K Context — Best Sub-100M Retrieval Quality
- Training and Finetuning Multimodal Embedding & Reranker Models with Sentence Transformers