← Back to community benchmarks

Qwen3-Coder-30B-A3B-Instruct

M5 Pro (20c) · 64 GB · 4bit · 2026-07-04
Performance
32k
tokens
1,072
PP tok/s
40.3
TG tok/s
30572
TTFT (ms)
19.6
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.4.4
macOS macOS 26.5.2
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
8k 1,610 56.7 17.5 GB view
32k 1,072 40.3 19.6 GB current