← Back to community benchmarks

Qwen3.6-35B-A3B-UD

M4 Pro (20c) · 48 GB · 4bit · 2026-04-23
Performance
16k
tokens
823.5
PP tok/s
54.9
TG tok/s
19895
TTFT (ms)
21.9
Peak mem (GB)
Hardware
Chip M4 Pro (20c)
Memory 48 GB
GPU Cores 20
Software
oMLX v0.3.6
macOS macOS 26.5
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 774.3 68.0 20.2 GB view
4k 904.0 65.1 21.0 GB view
8k 885.3 61.3 21.3 GB view
16k 823.5 54.9 21.9 GB current
32k 708.3 45.8 23.3 GB view
Batching Results
Batch Size TG tok/s Speedup
68.0 1.00×
105.5 1.55×