← Back to community benchmarks

Qwen3-Coder-30B-A3B-Instruct5bit

M1 Pro (16c) · 32 GB · 5bit · 2026-05-28
Performance
1k
tokens
160.8
PP tok/s
34.3
TG tok/s
6367
TTFT (ms)
19.8
Peak mem (GB)
Hardware
Chip M1 Pro (16c)
Memory 32 GB
GPU Cores 16
Software
oMLX v0.3.12
macOS macOS 15.7.4
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 160.8 34.3 19.8 GB current
4k 163.3 28.0 20.2 GB view
Batching Results
Batch Size TG tok/s Speedup
34.3 1.00×
67.9 1.98×
131.4 3.83×