A focused, no-frills embedding specialist that converts text into dense vector representations with strong semantic understanding. It handles similarity search, clustering, and retrieval tasks reliably, producing large 1024-dimensional embeddings that capture nuanced meaning. Open-weight and MIT-licensed, it's straightforward to deploy locally without usage restrictions.