← Back to community benchmarks

Qwen3.6-35B-A3B-oQ4e-fp16-mtp

M1 Max (32c) · 64 GB · 4bit · 2026-07-23

Performance

16k

tokens

803.8

PP tok/s

53.9

TG tok/s

20383

TTFT (ms)

21.3

Peak mem (GB)

Hardware

Chip M1 Max (32c)

Memory 64 GB

GPU Cores 32

Software

oMLX v0.5.3

macOS macOS 26.5.2

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	809.2	59.6	20.0 GB	view
4k	972.6	58.3	20.8 GB	view
8k	876.2	56.0	21.0 GB	view
16k	803.8	53.9	21.3 GB	current
32k	664.9	49.6	21.9 GB	view
64k	493.8	42.2	23.3 GB	view
128k	330.9	34.6	25.1 GB	view