A compact, no-frills embedding specialist that punches above its weight for its size. It converts text into dense vector representations efficiently, making it a practical choice for semantic search and similarity tasks. The 512-token context limit means it works best with shorter passages — long documents need chunking before it can process them meaningfully.