← Back to community benchmarks

Qwen3.6-35B-A3B-oQ6

M1 Max (32c) · 64 GB · 6bit · 2026-04-24
Performance
32k
tokens
791.5
PP tok/s
33.9
TG tok/s
41403
TTFT (ms)
31.0
Peak mem (GB)
Hardware
Chip M1 Max (32c)
Memory 64 GB
GPU Cores 32
Software
oMLX v0.3.7
macOS macOS 26.2
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 667.2 55.5 27.9 GB view
4k 947.2 53.4 28.7 GB view
8k 957.4 50.7 29.0 GB view
16k 906.3 45.6 29.7 GB view
32k 791.5 33.9 31.0 GB current