← Back to community benchmarks

Qwen3.6-35B-A3B-UD

M5 Pro (20c) · 64 GB · 4bit · 2026-04-28
Performance
16k
tokens
2,141
PP tok/s
65.1
TG tok/s
7654
TTFT (ms)
22.2
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.3.7
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,226 76.5 20.4 GB view
4k 1,966 74.0 21.2 GB view
8k 2,253 70.3 21.5 GB view
16k 2,141 65.1 22.2 GB current
Batching Results
Batch Size TG tok/s Speedup
76.5 1.00×
118.0 1.54×
165.9 2.17×