← Back to community benchmarks

Qwen3.5-35B-A3B

M3 Max (40c) · 48 GB · 4bit · 2026-05-04

Performance

32k

tokens

1,135

PP tok/s

72.3

TG tok/s

28880

TTFT (ms)

22.7

Peak mem (GB)

Hardware

Chip M3 Max (40c)

Memory 48 GB

GPU Cores 40

Software

oMLX v0.3.8

macOS macOS 26.3.1

Context 32,768

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
4k	1,463	89.3	20.0 GB	view
16k	1,366	79.5	21.0 GB	view
32k	1,135	72.3	22.7 GB	current