← Back to community benchmarks

Qwen3-Coder-Next

M3 Ultra (80c) · 512 GB · 4bit · 2026-03-14

Performance

8k

tokens

1,929

PP tok/s

65.0

TG tok/s

4248

TTFT (ms)

43.4

Peak mem (GB)

Hardware

Chip M3 Ultra (80c)

Memory 512 GB

GPU Cores 80

Software

oMLX v0.2.11

macOS macOS 15.3.2

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,561	71.7	43.0 GB	view
4k	1,919	68.4	43.2 GB	view
8k	1,929	65.0	43.4 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	71.7	1.00×
2×	125.0	1.74×