Bonsai 1.7B ternary model at 442T/s on M4 Max

by Rogelio Beer

Visit

May 4, 2026 AI & Machine Learning

autonomous search metal kernels model optimization

Gallery

About

We took a recently released Bonsai 1.7B ternary model from PrismML (https://github.com/PrismML-Eng/Bonsai-demo) and ran our agentic evolution search on it for 6 hours to optimize the Metal kernels. The search was fully autonomous.Measured against unmodified upstream llama.cpp at the same Bonsai/Q2_0 commit, same M4 Max:- tg128: 309.82 → 442.42 t/s (+42.0%)- pp512: 4250.32 → 4622.63 t/s (+8.8%)