← Back to community benchmarks

Qwen3.5-9B

M1 Max (24c) · 32 GB · 4bit · 2026-05-05

Performance

8k

tokens

238.1

PP tok/s

41.6

TG tok/s

34405

TTFT (ms)

6.9

Peak mem (GB)

Hardware

Chip M1 Max (24c)

Memory 32 GB

GPU Cores 24

Software

oMLX v0.3.7

macOS macOS 26.4.1

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	227.7	47.4	5.8 GB	view
4k	238.6	44.7	6.5 GB	view
8k	238.1	41.6	6.9 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	47.4	1.00×
2×	53.0	1.12×
4×	56.8	1.20×