mteb and mleb
MLEB is a specialized legal-domain benchmark that extends the evaluation methodology of MTEB to a specific corpus, making them complementary tools where MLEB users would typically also use MTEB for cross-domain baseline comparisons.
About mteb
embeddings-benchmark/mteb
MTEB: Massive Text Embedding Benchmark
This tool helps machine learning engineers and researchers assess the quality and performance of different text embedding models. You provide a text embedding model and specific evaluation tasks (like text classification or retrieval). The output is a clear set of metrics showing how well the model performs on those tasks, allowing for informed comparison and selection of the best model.
About mleb
isaacus-dev/mleb
The code used to evaluate embedding models on the Massive Legal Embedding Benchmark (MLEB).
This tool helps legal tech developers, researchers, and data scientists evaluate how well their legal text embedding models understand and reason about legal documents. You input your embedding model and relevant API keys, and it outputs a performance score across diverse legal datasets. This helps you understand your model's strengths and weaknesses in legal applications.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work