Show HN: Needle2: 14MB agentic LLM for phones, wearables, smart home and robots
Summary
Needle2 is an on-device 45M-parameter LLM optimized for tool calling, device interaction, and structured extraction, packaged as a 14MB binary that runs in about 28MB of RAM. It targets wearables, phones, smart homes, and robots, with benchmarks showing strong on-device performance and energy efficiency across mobile, VR, and low-cost devices, and it is released under Apache 2.0 with on-device tooling and testing.