← Back to community benchmarks

Qwen3-4B-Instruct-2507

M2 (10c) · 16 GB · 4bit · 2026-03-11

Performance

1k

tokens

194.1

PP tok/s

31.1

TG tok/s

5275

TTFT (ms)

2.8

Peak mem (GB)

Hardware

Chip M2 (10c)

Memory 16 GB

GPU Cores 10

Software

oMLX v0.2.7

macOS macOS 26.2

Context 1,024

Batching Results

Batch Size	TG tok/s	Speedup
1×	31.1	1.00×
2×	42.9	1.38×