← Back to community benchmarks

Qwen3.6-35B-A3B-oQ2

M5 (10c) · 32 GB · 2bit · 2026-04-30
Performance
64k
tokens
956.4
PP tok/s
13.1
TG tok/s
68524
TTFT (ms)
16.4
Peak mem (GB)
Hardware
Chip M5 (10c)
Memory 32 GB
GPU Cores 10
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 65,536
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 574.8 41.2 12.8 GB view
4k 696.7 40.3 13.5 GB view
8k 719.4 37.7 13.8 GB view
16k 1,360 14.4 14.9 GB view
32k 1,166 12.9 15.5 GB view
64k 956.4 13.1 16.4 GB current
Batching Results
Batch Size TG tok/s Speedup
41.2 1.00×
53.3 1.29×
66.3 1.61×