← Back to community benchmarks
Qwen3.6-35B-A3B-RotorQuant
Performance
1k
tokens
729.4
PP tok/s
89.5
TG tok/s
1404
TTFT (ms)
11.1
Peak mem (GB)
Hardware
Chip
M5 (10c)
Memory
32 GB
GPU Cores
10
Software
oMLX
v0.3.6
macOS
macOS 26.4.1
Context
1,024