← Back to community benchmarks

Jundot-Qwen3.6-27B-oQ4-mtp

M5 Max (40c) · 128 GB · 4bit · 2026-05-22
Performance
4k
tokens
826.5
PP tok/s
45.1
TG tok/s
4956
TTFT (ms)
18.2
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.9
macOS macOS 26.5
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 766.4 45.7 16.7 GB view
4k 826.5 45.1 18.2 GB current
195k 304.7 17.5 48.6 GB view
Batching Results
Batch Size TG tok/s Speedup
45.7 1.00×
50.9 1.11×
58.1 1.27×