← Back to community benchmarks

Qwen3.6-35B-A3B-UD

M1 Max (32c) · 64 GB · 4bit · 2026-04-25

Performance

16k

tokens

453.8

PP tok/s

41.3

TG tok/s

36102

TTFT (ms)

22.2

Peak mem (GB)

Hardware

Chip M1 Max (32c)

Memory 64 GB

GPU Cores 32

Software

oMLX v0.3.7

macOS macOS 26.3

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	461.4	93.7	21.1 GB	view
4k	365.9	53.2	21.2 GB	view
8k	528.2	44.8	21.5 GB	view
16k	453.8	41.3	22.2 GB	current
32k	415.2	34.0	23.5 GB	view
64k	354.0	23.7	26.2 GB	view
128k	216.7	7.0	31.6 GB	view