Base Compute gives AI agents the kernel-tuning job for open models

Its B:OS demo took Nemotron 3 Nano from unsupported to 1.43x faster decode than MLX on Apple silicon, by Base Compute's tests.

By · Published

A stylized silicon chip die with intricate micro-architectural patterns, photographed from a low angle on a seamless backdrop.

Its B:OS demo took Nemotron 3 Nano from unsupported to 1.43x faster decode than MLX on Apple silicon, by Base Compute's tests.