← Back to community benchmarks

Qwen3-Coder-Next

M3 Max (30c) · 96 GB · 4bit · 2026-07-18

Performance

16k

tokens

681.4

PP tok/s

51.6

TG tok/s

24043

TTFT (ms)

44.3

Peak mem (GB)

Hardware

Chip M3 Max (30c)

Memory 96 GB

GPU Cores 30

Software

oMLX v0.5.1

macOS macOS 26.6

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
4k	720.6	54.0	43.8 GB	view
16k	681.4	51.6	44.3 GB	current
32k	613.3	47.3	45.1 GB	view
64k	479.0	36.8	46.2 GB	view