← Back to community benchmarks

Qwen3-Coder-30B-A3B-Instruct5bit

M5 Pro (20c) · 48 GB · 5bit · 2026-05-11

Performance

1k

tokens

1,134

PP tok/s

73.4

TG tok/s

903

TTFT (ms)

20.1

Peak mem (GB)

Hardware

Chip M5 Pro (20c)

Memory 48 GB

GPU Cores 20

Software

oMLX v0.3.9.dev1

macOS macOS 26.4.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,134	73.4	20.1 GB	current
4k	2,183	66.1	20.8 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	73.4	1.00×
2×	106.6	1.45×
4×	129.0	1.76×