← Back to community benchmarks

Qwen3-Coder-Next

M1 Max (24c) · 64 GB · 4bit · 2026-03-12

Performance

8k

tokens

261.4

PP tok/s

37.6

TG tok/s

31338

TTFT (ms)

43.4

Peak mem (GB)

Hardware

Chip M1 Max (24c)

Memory 64 GB

GPU Cores 24

Software

oMLX v0.2.9

macOS macOS 26.3.1

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	260.3	34.8	43.0 GB	view
4k	266.5	40.4	43.2 GB	view
8k	261.4	37.6	43.4 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	34.8	1.00×
2×	55.1	1.58×
4×	60.0	1.72×