You can now run Liquid AI's LFM3 models, built for tool useA model's ability to call external functions — run code, search the web, edit files — instead of only generating text.Full definition → and structured outputForcing a model's response to match a schema, so downstream code can parse it instead of guessing at prose.Full definition →, locally with one command on Apple hardware. The release also speeds up prompt processing and adds DFlash2 speculative decodingA speed trick where a small model drafts several tokens ahead and the big model verifies them in one pass, often doubling generation speed.Full definition → on MLX.
Terms in this piece · Glossary
speculative decoding — A speed trick where a small model drafts several tokens ahead and the big model verifies them in one pass, often doubling generation speed.
tool use — A model's ability to call external functions — run code, search the web, edit files — instead of only generating text.
structured output — Forcing a model's response to match a schema, so downstream code can parse it instead of guessing at prose.
