← Back to community benchmarks

Qwen3.5-9B

M4 (10c) · 16 GB · 4bit · 2026-03-11

Performance

4k

tokens

137.6

PP tok/s

15.6

TG tok/s

29758

TTFT (ms)

6.8

Peak mem (GB)

Hardware

Chip M4 (10c)

Memory 16 GB

GPU Cores 10

Software

oMLX v0.2.7

macOS macOS 15.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	140.3	16.0	6.5 GB	view
4k	137.6	15.6	6.8 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	16.0	1.00×
2×	25.4	1.59×
4×	29.9	1.87×