Benchmark
MTEB
Compare embedding models for search, classification, clustering, and related tasks.
Published leaderboard
Explore scores and evaluation settings on the publisher’s leaderboard.
Open leaderboardCoverage
Embedding and retrieval evaluations across tasks, languages, and modalities. The original text benchmark and later multilingual or multimodal suites have distinct scopes.
Methodology
Evaluate models on a named benchmark configuration and compare task-specific metrics or the matching leaderboard aggregate.