Launch HN: Magnitude (YC S25) – Self-optimizing inference engine for agents

Magnitude is an open-source inference engine for AI agents that optimizes itself for individual hardware, allowing open models to run up to 2x faster than llama.cpp by compiling and tuning kernels on your device. It works as a desktop app across Apple Silicon, NVIDIA, AMD, and CPU-only systems, and integrates with popular agents like Pi and OpenCode with a single click. The platform is free, private, and uses no external tokens, keeping all processing local.