← Back to community benchmarks

Qwen3.6-35B-A3B

M4 Pro (20c) · 64 GB · 4bit · 2026-04-16

Performance

64k

tokens

504.3

PP tok/s

42.9

TG tok/s

129957

TTFT (ms)

25.0

Peak mem (GB)

Hardware

Chip M4 Pro (20c)

Memory 64 GB

GPU Cores 20

Software

oMLX v0.3.5

macOS macOS 26.3

Context 65,536

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	734.9	76.7	19.2 GB	view
4k	889.5	74.1	20.0 GB	view
8k	834.4	71.0	20.4 GB	view
16k	703.6	64.7	21.0 GB	view
32k	610.5	57.3	22.4 GB	view
64k	504.3	42.9	25.0 GB	current