← Back to community benchmarks

Qwen3.6-35B-A3B

M4 Pro (16c) · 48 GB · 4bit · 2026-04-25

Performance

64k

tokens

449.9

PP tok/s

25.4

TG tok/s

145679

TTFT (ms)

25.0

Peak mem (GB)

Hardware

Chip M4 Pro (16c)

Memory 48 GB

GPU Cores 16

Software

oMLX v0.3.6

macOS macOS 26.4.1

Context 65,536

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	586.6	168.0	20.6 GB	view
4k	452.9	74.9	20.0 GB	view
8k	725.0	73.6	20.4 GB	view
16k	679.4	63.7	21.0 GB	view
32k	585.5	46.0	22.4 GB	view
64k	449.9	25.4	25.0 GB	current