← Back to community benchmarks

Qwen3.6-35B-A3B-oQ6e-fp16-mtp

M2 Ultra (76c) · 192 GB · 6bit · 2026-07-19
User commentv0.5.1 5a39ba3a with custom kernels
Performance
8k
tokens
2,334
PP tok/s
81.3
TG tok/s
3510
TTFT (ms)
28.7
Peak mem (GB)
Hardware
Chip M2 Ultra (76c)
Memory 192 GB
GPU Cores 76
Software
oMLX v0.5.1
macOS macOS 26.5.2
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,362 83.5 27.8 GB view
4k 2,522 82.9 28.5 GB view
8k 2,334 81.3 28.7 GB current
16k 2,055 78.5 29.1 GB view
32k 1,719 74.0 29.4 GB view
64k 1,270 66.2 31.0 GB view
128k 830.5 56.1 33.1 GB view