← Back to community benchmarks

ThinkingCap-Qwen3.6-27B-OptiQ

M3 Ultra (60c) · 256 GB · 4bit · 2026-07-22
Performance
4k
tokens
331.3
PP tok/s
31.7
TG tok/s
12364
TTFT (ms)
20.4
Peak mem (GB)
Hardware
Chip M3 Ultra (60c)
Memory 256 GB
GPU Cores 60
Software
oMLX v0.5.3
macOS macOS 26.5.2
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 324.3 30.5 19.2 GB view
4k 331.3 31.7 20.4 GB current
8k 326.8 30.9 20.7 GB view
16k 314.9 29.6 21.2 GB view
32k 292.9 26.6 22.4 GB view
64k 256.7 24.7 24.8 GB view