← Back to community benchmarks

Qwen3.6-35B-A3B-UD

M3 Pro (14c) · 36 GB · 4bit · 2026-04-25

Performance

4k

tokens

551.6

PP tok/s

37.2

TG tok/s

7425

TTFT (ms)

21.2

Peak mem (GB)

Hardware

Chip M3 Pro (14c)

Memory 36 GB

GPU Cores 14

Software

oMLX v0.3.7rc2

macOS macOS 15.7.5

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	497.7	39.2	20.4 GB	view
4k	551.6	37.2	21.2 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	39.2	1.00×
2×	63.4	1.62×
4×	85.1	2.17×