← Back to community benchmarks

Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive-OptiQ-5bpw

M5 Pro (20c) · 64 GB · 4bit · 2026-07-21
Performance
4k
tokens
2,438
PP tok/s
85.8
TG tok/s
1680
TTFT (ms)
20.3
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.5.1
macOS macOS 26.4
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,929 88.1 19.5 GB view
4k 2,438 85.8 20.3 GB current
16k 2,149 77.8 21.4 GB view
32k 1,772 69.0 23.0 GB view
64k 1,290 56.6 26.0 GB view
128k 818.1 42.1 31.3 GB view
195k 562.1 33.0 35.0 GB view
Batching Results
Batch Size TG tok/s Speedup
88.1 1.00×
86.0 0.98×
186.0 2.11×
423.1 4.80×