← Back to community benchmarks

Qwen3.5-0.8B

M4 Pro (20c) · 24 GB · bf16 · 2026-03-19

Performance

32k

tokens

3,224

PP tok/s

84.7

TG tok/s

10165

TTFT (ms)

5.1

Peak mem (GB)

Hardware

Chip M4 Pro (20c)

Memory 24 GB

GPU Cores 20

Software

oMLX v0.2.19

macOS macOS 26.3.1

Context 32,768

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	4,624	130.9	2.5 GB	view
4k	4,553	122.9	2.9 GB	view
8k	4,379	108.9	3.2 GB	view
16k	3,924	103.5	3.9 GB	view
32k	3,224	84.7	5.1 GB	current
64k	2,350	64.0	7.6 GB	view