← Back to community benchmarks

Qwen3.6-35B-A3B-5bit

M4 Pro (20c) · 48 GB · 5bit · 2026-04-30

Performance

4k

tokens

872.5

PP tok/s

70.1

TG tok/s

4695

TTFT (ms)

24.1

Peak mem (GB)

Hardware

Chip M4 Pro (20c)

Memory 48 GB

GPU Cores 20

Software

oMLX v0.3.8rc1

macOS macOS 26.3.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	703.5	73.2	23.3 GB	view
4k	872.5	70.1	24.1 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	73.2	1.00×
2×	101.5	1.39×
4×	118.9	1.62×