← Back to community benchmarks

Qwen3.5-2B

M4 (10c) · 24 GB · 4bit · 2026-03-10

Performance

1k

tokens

1,009

PP tok/s

70.8

TG tok/s

1015

TTFT (ms)

2.5

Peak mem (GB)

Hardware

Chip M4 (10c)

Memory 24 GB

GPU Cores 10

Software

oMLX v0.2.6

macOS macOS 26.0.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,009	70.8	2.5 GB	current
4k	1,031	75.0	2.5 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	70.8	1.00×
2×	149.9	2.12×
4×	181.4	2.56×