← Back to community benchmarks

Qwen3.5-9B

M1 Max (24c) · 32 GB · bf16 · 2026-05-04

Performance

8k

tokens

380.7

PP tok/s

18.6

TG tok/s

21521

TTFT (ms)

18.7

Peak mem (GB)

Hardware

Chip M1 Max (24c)

Memory 32 GB

GPU Cores 24

Software

oMLX v0.3.8

macOS macOS 26.3

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	300.1	72.4	19.9 GB	view
4k	240.3	19.0	18.3 GB	view
8k	380.7	18.6	18.7 GB	current
16k	356.5	17.7	19.8 GB	view
32k	334.9	16.1	20.9 GB	view