← Back to community benchmarks

Qwen3.5-35B-A3B

M4 Pro (20c) · 48 GB · 4bit · 2026-03-10

Performance

64k

tokens

515.1

PP tok/s

22.8

TG tok/s

127238

TTFT (ms)

23.3

Peak mem (GB)

Hardware

Chip M4 Pro (20c)

Memory 48 GB

GPU Cores 20

Software

oMLX v0.2.6

macOS macOS 26.3

Context 65,536

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	880.1	82.8	20.1 GB	view
4k	901.9	80.8	20.2 GB	view
8k	866.0	63.6	20.4 GB	view
16k	798.4	63.6	20.8 GB	view
32k	686.9	49.6	21.6 GB	view
64k	515.1	22.8	23.3 GB	current