← Back to community benchmarks

Qwen3.5-35B-A3B

M4 Max (40c) · 64 GB · 4bit · 2026-04-14

Performance

16k

tokens

1,550

PP tok/s

71.3

TG tok/s

10573

TTFT (ms)

39.2

Peak mem (GB)

Hardware

Chip M4 Max (40c)

Memory 64 GB

GPU Cores 40

Software

oMLX v0.3.5

macOS macOS 26.4.1

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,461	366.1	38.1 GB	view
4k	1,327	84.8	38.4 GB	view
8k	1,656	76.0	38.6 GB	view
16k	1,550	71.3	39.2 GB	current
32k	1,242	50.5	40.4 GB	view
64k	972.4	31.8	42.8 GB	view
128k	614.4	19.7	47.6 GB	view