Liquid AI Ships LFM2.5-230M for On-Device Inference

Liquid AI released LFM2.5-230M, its smallest 230-million-parameter model for agentic tasks on phones and robots. The open-weight model supports llama.cpp, MLX, vLLM, SGLang, and ONNX. It scores 71.71 on IFEval, beating Qwen3.5-0.8B and Gemma 3 1B IT.