← Back to community benchmarks

Qwen3.6-35B-A3B-oQ6

M4 Pro (20c) · 64 GB · 6bit · 2026-05-04
Performance
16k
tokens
742.5
PP tok/s
47.2
TG tok/s
22067
TTFT (ms)
29.4
Peak mem (GB)
Hardware
Chip M4 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.3.8
macOS macOS 26.3
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 729.1 61.7 27.7 GB view
4k 880.1 58.9 28.4 GB view
8k 856.4 53.8 28.8 GB view
16k 742.5 47.2 29.4 GB current
32k 617.9 42.4 30.8 GB view
Batching Results
Batch Size TG tok/s Speedup
61.7 1.00×
92.1 1.49×