← Back to community benchmarks

Qwen3.6-35B-A3B-oQ2

M4 Pro (16c) · 24 GB · 2bit · 2026-04-17
Performance
4k
tokens
745.7
PP tok/s
70.5
TG tok/s
5493
TTFT (ms)
13.3
Peak mem (GB)
Hardware
Chip M4 Pro (16c)
Memory 24 GB
GPU Cores 16
Software
oMLX v0.3.6
macOS macOS 26.4.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 657.2 75.1 12.6 GB view
4k 745.7 70.5 13.3 GB current
8k 731.9 66.0 13.6 GB view
Batching Results
Batch Size TG tok/s Speedup
75.1 1.00×
107.0 1.42×
125.7 1.67×