← Back to community benchmarks

Qwen3.6-35B-A3B-OptiQ

M5 Pro (20c) · 64 GB · 4bit · 2026-06-08
Performance
4k
tokens
2,087
PP tok/s
71.4
TG tok/s
1963
TTFT (ms)
22.5
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.4.1
macOS macOS 26.5.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,383 70.7 21.7 GB view
4k 2,087 71.4 22.5 GB current
Batching Results
Batch Size TG tok/s Speedup
70.7 1.00×
116.1 1.64×
229.1 3.24×