← Back to community benchmarks

Qwen3.6-35B-A3B-UD

M1 Ultra (64c) · 128 GB · 4bit · 2026-05-18
Performance
16k
tokens
1,093
PP tok/s
56.3
TG tok/s
14997
TTFT (ms)
22.2
Peak mem (GB)
Hardware
Chip M1 Ultra (64c)
Memory 128 GB
GPU Cores 64
Software
oMLX v0.3.8
macOS macOS 26.3
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 847.6 66.2 20.4 GB view
4k 1,102 63.7 21.2 GB view
8k 1,119 60.8 21.7 GB view
16k 1,093 56.3 22.2 GB current
Batching Results
Batch Size TG tok/s Speedup
66.2 1.00×
107.9 1.63×
163.3 2.47×