all-minilm-l6-v2

all-MiniLM-L6-v2 is a lightweight sentence embedding model that converts sentences and short paragraphs into 384-dimensional dense vectors. These embeddings capture semantic meaning, making the model well suited for semantic search, retrieval, similarity scoring, clustering, deduplication, recommendation, and text ranking. Its compact architecture is optimized for fast, low-cost inference, making it a strong choice for high-volume embedding workloads and retrieval pipelines.