onil_gova

M5 Max vs M3 Max: Local LLM Benchmarks

Side-by-side Qwen 3.5 inference benchmarks on M5 Max and M3 Max MacBook Pros: token generation, prefill speed and batching.