MTEB
The Massive Text Embedding Benchmark, a public set of tasks and a Hugging Face leaderboard for comparing embedding models.
In more detail
Use the retrieval scores to build a shortlist, then test the shortlist on your own labelled queries: leaderboard tasks rarely match your text, and some models are tuned to the benchmark. Scores also depend on the language and domain of each task, so read the per-task results.
Related terms
- embedding model — A model that maps text to embedding vectors rather than generating text.
- leaderboard — A public ranking of models by their scores on shared benchmark tasks.
Taught in
- Embedding Models — AI Engineering: LLM Apps, RAG & Agents