← Back to community benchmarks

Qwen3.6-35B-A3B8bit

M2 Max (38c) · 96 GB · 8bit · 2026-05-04
Performance
8k
tokens
1,337
PP tok/s
65.7
TG tok/s
6126
TTFT (ms)
36.5
Peak mem (GB)
Hardware
Chip M2 Max (38c)
Memory 96 GB
GPU Cores 38
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,071 70.1 35.4 GB view
4k 1,339 67.7 36.2 GB view
8k 1,337 65.7 36.5 GB current
16k 1,255 61.8 37.4 GB view
32k 1,089 56.2 38.5 GB view
64k 837.5 47.1 41.2 GB view
128k 542.6 36.4 46.5 GB view
195k 378.3 28.6 54.8 GB view