Devstral Small-2505 Q4_K_M; llama-bench TG128 / PP512
on M5 Max 48GB · llama.cpp
Summary
User reports Devstral-Small-24B at 37.6 t/s generation and 978.1 t/s prompt processing on a single M5 Max with 48GB unified memory.
Setup is llama.cpp build 11146 with Metal, Q4_K_M weights, n_gpu_layers=99 and flash_attention=true.
The figures are separate llama-bench microtests, tg128=37.56 t/s and pp512=978.12 t/s, not one 512-input/128-output request; context length is unspecified.