← Back to community benchmarks

Qwen3-Coder-30B-A3B-Instruct

M5 Pro (20c) · 64 GB · 4bit · 2026-03-23

Performance

1k

tokens

1,772

PP tok/s

90.4

TG tok/s

578

TTFT (ms)

16.6

Peak mem (GB)

Hardware

Chip M5 Pro (20c)

Memory 64 GB

GPU Cores 20

Software

oMLX v0.2.19

macOS macOS 26.3.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,772	90.4	16.6 GB	current
4k	2,027	73.9	17.1 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	90.4	1.00×
2×	178.9	1.98×
4×	358.5	3.97×