← Back to community benchmarks

Ministral-3-8B-Instruct-2512

M3 (10c) · 24 GB · 4bit · 2026-03-16

Performance

1k

tokens

177.7

PP tok/s

19.2

TG tok/s

5762

TTFT (ms)

5.9

Peak mem (GB)

Hardware

Chip M3 (10c)

Memory 24 GB

GPU Cores 10

Software

oMLX v0.2.13

macOS macOS 26.2

Context 1,024

Batching Results

Batch Size	TG tok/s	Speedup
1×	19.2	1.00×
2×	36.4	1.90×