← Back to community benchmarks

Qwen3.5-9B-DeepSeek-V4-Flash

M5 Pro (20c) · 64 GB · 8bit · 2026-05-18

Performance

4k

tokens

1,463

PP tok/s

30.7

TG tok/s

2799

TTFT (ms)

10.6

Peak mem (GB)

Hardware

Chip M5 Pro (20c)

Memory 64 GB

GPU Cores 20

Software

oMLX v0.3.9.dev2

macOS macOS 26.5

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,146	31.3	9.9 GB	view
4k	1,463	30.7	10.6 GB	current
8k	1,504	30.8	11.2 GB	view
16k	1,431	29.5	12.0 GB	view