← Back to community benchmarks

Qwen3.5-35B-A3B

M5 Pro (20c) · 64 GB · 8bit · 2026-03-26

Performance

64k

tokens

1,253

PP tok/s

23.8

TG tok/s

52289

TTFT (ms)

42.1

Peak mem (GB)

Hardware

Chip M5 Pro (20c)

Memory 64 GB

GPU Cores 20

Software

oMLX v0.2.21

macOS macOS 26.3.1

Context 65,536

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,535	49.5	36.2 GB	view
4k	2,151	47.5	37.1 GB	view
8k	2,168	44.7	37.5 GB	view
16k	2,031	40.2	38.2 GB	view
32k	1,727	31.9	39.5 GB	view
64k	1,253	23.8	42.1 GB	current
128k	801.7	15.5	47.5 GB	view