CPUs Aren't Dead. Gemma2B Out Scored GPT-3.5 Turbo on Test That Made It Famous

Create an optimized inference library specifically for CPU-bound small language models using modern instruction sets (AVX-512/AMX). Focus on reducing latency for models under 3B parameters.

Suggested repo: cpu-infer

"Stop wasting GPU cycles on small model tasks."

Estimated effort: 40h