← Back to community benchmarks

Qwen3.5-9B

M5 (8c) · 16 GB · 4bit · 2026-04-18

Performance

1k

tokens

609.3

PP tok/s

24.9

TG tok/s

1681

TTFT (ms)

6.5

Peak mem (GB)

Hardware

Chip M5 (8c)

Memory 16 GB

GPU Cores 8

Software

oMLX v0.2.22

macOS macOS 26.4.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	609.3	24.9	6.5 GB	current
4k	584.4	24.3	7.3 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	24.9	1.00×
2×	43.9	1.76×
4×	42.6	1.71×