← Back to community benchmarks

Qwen3.6-35B-A3B-oQ6

M5 Pro (20c) · 64 GB · 6bit · 2026-04-18
Performance
1k
tokens
1,216
PP tok/s
70.7
TG tok/s
842
TTFT (ms)
27.7
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.3.6
macOS macOS 26.4.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,216 70.7 27.7 GB current
4k 1,941 69.6 28.5 GB view
8k 2,115 65.4 28.9 GB view
Batching Results
Batch Size TG tok/s Speedup
70.7 1.00×
107.6 1.52×
149.5 2.11×
181.1 2.56×