← Back to community benchmarks

Qwen3.6-27B-oQ8-mtp

M3 Ultra (80c) · 96 GB · 8bit · 2026-05-13
Performance
32k
tokens
385.1
PP tok/s
29.1
TG tok/s
85091
TTFT (ms)
35.8
Peak mem (GB)
Hardware
Chip M3 Ultra (80c)
Memory 96 GB
GPU Cores 80
Software
oMLX v0.3.9.dev1
macOS macOS 26.5
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 388.4 30.8 28.8 GB view
4k 420.9 33.6 30.3 GB view
8k 420.3 32.7 31.3 GB view
16k 409.5 31.3 32.8 GB view
32k 385.1 29.1 35.8 GB current
64k 335.0 25.6 41.8 GB view
Batching Results
Batch Size TG tok/s Speedup
30.8 1.00×
40.2 1.31×
61.8 2.01×