← Back to community benchmarks

Qwen3.6-35B-A3B-oQ6-mtp

M5 Pro (20c) · 64 GB · 6bit · 2026-05-18
Performance
1k
tokens
1,413
PP tok/s
73.8
TG tok/s
725
TTFT (ms)
28.3
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.3.9.dev2
macOS macOS 26.4.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,413 73.8 28.3 GB current
4k 1,684 71.1 29.1 GB view
8k 1,806 67.6 29.4 GB view
16k 1,781 65.7 30.3 GB view
32k 1,568 59.1 31.8 GB view
Batching Results
Batch Size TG tok/s Speedup
73.8 1.00×
105.4 1.43×