← Back to community benchmarks

Qwen3-Coder-30B-A3B-Instruct

M1 Max (32c) · 64 GB · 4bit · 2026-05-18
Performance
8k
tokens
213.9
PP tok/s
28.1
TG tok/s
38290
TTFT (ms)
17.8
Peak mem (GB)
Hardware
Chip M1 Max (32c)
Memory 64 GB
GPU Cores 32
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 238.0 43.2 16.6 GB view
4k 239.6 35.3 17.2 GB view
8k 213.9 28.1 17.8 GB current
Batching Results
Batch Size TG tok/s Speedup
43.2 1.00×
52.0 1.20×
53.5 1.24×