← Back to community benchmarks

Ministral-3-8B-Instruct-2512

M1 Pro (14c) · 16 GB · 4bit · 2026-03-10

Performance

1k

tokens

130.1

PP tok/s

28.9

TG tok/s

7873

TTFT (ms)

6.0

Peak mem (GB)

Hardware

Chip M1 Pro (14c)

Memory 16 GB

GPU Cores 14

Software

oMLX v0.2.7

macOS macOS 26.3.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	130.1	28.9	6.0 GB	current
4k	124.7	24.4	6.4 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	28.9	1.00×
2×	28.8	1.00×