← Back to community benchmarks

Qwen3.6-35B-A3B-oQ6-fp16-mtp

M4 Pro (20c) · 48 GB · 6bit · 2026-05-09
Performance
32k
tokens
737.5
PP tok/s
49.3
TG tok/s
44429
TTFT (ms)
31.8
Peak mem (GB)
Hardware
Chip M4 Pro (20c)
Memory 48 GB
GPU Cores 20
Software
oMLX v0.3.9.dev1
macOS macOS 26.4.1
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 759.9 63.9 28.3 GB view
4k 941.3 63.0 29.1 GB view
8k 927.1 59.9 29.4 GB view
16k 861.3 56.3 30.3 GB view
32k 737.5 49.3 31.8 GB current
Batching Results
Batch Size TG tok/s Speedup
63.9 1.00×
94.0 1.47×
114.8 1.80×
125.9 1.97×