Mteb
/All repositories
Overview
HEALTH SCORE
Healthy
83out of 100
Solid maintainer coverage and development cadence. Security posture keeps the score from reaching Excellent.
LIFECYCLE
Active
Consistent commits, responsive maintainers, regular releases, and healthy issue triage.
Health breakdown
Healthy(83/100)
Maintainer availability, known risks and supply chain posture, and whether the project is actively developed or appropriately stable.
Maintainer responsiveness
Median response time is 0 days.
Bus factor
35 active maintainers currently have merge rights.
Company diversity
Contributors span 121 companies.
Project Achievements
LeaderboardsAbout
MTEB is a comprehensive evaluation framework for embeddings and retrieval systems across multiple languages and modalities. It provides standardized benchmarks for assessing the performance of embedding models on tasks including semantic search, clustering, information retrieval, text classification, bitext mining, and reranking. The project includes an interactive leaderboard and tools for running evaluations via both Python API and command-line interface.
First commit
Software value
$73M
Tags
benchmark
bitext-mining
clustering
embeddings
evaluation
information-retrieval
low-resource-nlp
mteb
multilingual-nlp
multimodal
neural-search
reranking
retrieval
sbert
semantic-search
sentence-transformers
sts
text-classification
text-embedding
License(s)
Apache-2.0
Data sources
Git
GitHub