← Back to community benchmarks

Qwen3.6-35B-A3B-TurboQuant

M5 Max (40c) · 128 GB · 8bit · 2026-04-17

Performance

32k

tokens

3,166

PP tok/s

74.0

TG tok/s

10349

TTFT (ms)

38.5

Peak mem (GB)

Hardware

Chip M5 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.6

macOS macOS 26.5

Context 32,768

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	2,140	98.0	35.4 GB	view
4k	3,585	95.1	36.2 GB	view
8k	3,832	91.7	36.5 GB	view
16k	3,577	85.8	37.1 GB	view
32k	3,166	74.0	38.5 GB	current
64k	2,294	56.3	41.2 GB	view
128k	1,315	37.3	46.5 GB	view