← Back to community benchmarks

Qwen3.6-35B-A3B-UD

M5 Pro (20c) · 48 GB · 4bit · 2026-07-05

Performance

4k

tokens

2,067

PP tok/s

71.7

TG tok/s

1981

TTFT (ms)

21.2

Peak mem (GB)

Hardware

Chip M5 Pro (20c)

Memory 48 GB

GPU Cores 20

Software

oMLX v0.4.5.dev1

macOS macOS 26.5.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,397	73.2	20.4 GB	view
4k	2,067	71.7	21.2 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	73.2	1.00×
2×	114.5	1.56×
4×	156.0	2.13×