← Back to community benchmarks

unsloth--Qwen3.6-35B-A3B-UD

M5 Pro (20c) · 64 GB · 4bit · 2026-07-18
Performance
32k
tokens
1,011
PP tok/s
59.7
TG tok/s
32420
TTFT (ms)
22.4
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.5.1
macOS macOS 26.5.2
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,004 73.7 20.4 GB view
4k 2,430 72.3 21.2 GB view
8k 2,363 69.5 21.5 GB view
16k 1,501 65.9 21.7 GB view
32k 1,011 59.7 22.4 GB current
64k 624.1 49.9 22.7 GB view