← Back to community benchmarks

Qwen3.6-27B-oQ4-mtp

M3 Ultra (80c) · 512 GB · 4bit · 2026-05-14
Performance
16k
tokens
405.8
PP tok/s
41.1
TG tok/s
40379
TTFT (ms)
20.7
Peak mem (GB)
Hardware
Chip M3 Ultra (80c)
Memory 512 GB
GPU Cores 80
Software
oMLX v0.3.9.dev1
macOS macOS 26.2
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 384.9 46.4 16.7 GB view
4k 417.3 45.3 18.2 GB view
8k 416.1 42.0 19.2 GB view
16k 405.8 41.1 20.7 GB current
32k 377.9 38.6 23.7 GB view
Batching Results
Batch Size TG tok/s Speedup
46.4 1.00×
60.1 1.30×
80.7 1.74×
92.7 2.00×