← Back to community benchmarks

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

M1 Max (24c) · 64 GB · 4bit · 2026-03-12
Performance
16k
tokens
70.7
PP tok/s
12.1
TG tok/s
231664
TTFT (ms)
16.5
Peak mem (GB)
Hardware
Chip M1 Max (24c)
Memory 64 GB
GPU Cores 24
Software
oMLX v0.2.7
macOS macOS 26.3.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 76.0 18.2 13.9 GB view
4k 75.6 16.5 14.6 GB view
8k 74.0 14.9 15.2 GB view
16k 70.7 12.1 16.5 GB current
32k 64.8 9.2 19.0 GB view
Batching Results
Batch Size TG tok/s Speedup
18.2 1.00×
18.3 1.01×
19.0 1.04×
17.1 0.94×