← Back to community benchmarks

Devstral-Small-2-24B-Instruct-2512

M5 Max (40c) · 128 GB · 4bit · 2026-05-18

Performance

4k

tokens

1,060

PP tok/s

37.4

TG tok/s

3864

TTFT (ms)

14.1

Peak mem (GB)

Hardware

Chip M5 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.9.dev2

macOS macOS 26.4.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	992.9	39.5	13.2 GB	view
4k	1,060	37.4	14.1 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	39.5	1.00×
2×	72.5	1.84×
4×	86.0	2.18×