← Back to community benchmarks

Qwen3___5-9B

M3 Max (40c) · 64 GB · 4bit · 2026-04-12

Performance

1k

tokens

665.5

PP tok/s

70.4

TG tok/s

1539

TTFT (ms)

5.8

Peak mem (GB)

Hardware

Chip M3 Max (40c)

Memory 64 GB

GPU Cores 40

Software

oMLX v0.3.5.dev1

macOS macOS 26.4.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	665.5	70.4	5.8 GB	current
4k	730.5	68.0	6.5 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	70.4	1.00×
2×	125.9	1.79×
4×	150.9	2.14×