Skip to content
llm-speed
Leaderboard/model/codestral-22b-v0-1-4bit

Codestral-22B-v0.1-4bit

2 workload results across 2 hardware configurations.

Fastest known config

47.5 decode tok/s

on M3 Ultra (60-core GPU) + 96GB unified via mlx see full run

M3 Pro (18-core GPU) + 36GB unifiedM3 Pro (18-core GPU) + 36GB unified

WorkloadBackendQuantdecode tok/sprefill tok/sTTFTRun
chat-shortmlx@0.31.3no datano datano datar_yyr1b4q9kx3

M3 Ultra (60-core GPU) + 96GB unifiedM3 Ultra (60-core GPU) + 96GB unified

WorkloadBackendQuantdecode tok/sprefill tok/sTTFTRun
chat-shortmlx@0.31.347.49tok/s205.7tok/s559msr_79dvtag5fd_

Codestral-22B-v0.1-4bit on hardware