← Back to community benchmarks

Qwen3.6-27B-oQ8-mtp

M2 Ultra (76c) · 128 GB · 8bit · 2026-05-07
Performance
4k
tokens
226.1
PP tok/s
29.3
TG tok/s
18117
TTFT (ms)
30.3
Peak mem (GB)
Hardware
Chip M2 Ultra (76c)
Memory 128 GB
GPU Cores 76
Software
oMLX v0.3.9.dev1
macOS macOS 15.7.5
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 210.1 30.8 28.8 GB view
4k 226.1 29.3 30.3 GB current
8k 226.6 28.2 31.3 GB view
Batching Results
Batch Size TG tok/s Speedup
30.8 1.00×
37.9 1.23×
51.8 1.68×