← Back to community benchmarks

Qwen3.6-35B-A3B-UD

M5 Max (40c) · 128 GB · 4bit · 2026-05-12
Performance
16k
tokens
1,884
PP tok/s
84.2
TG tok/s
8695
TTFT (ms)
22.4
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.9.dev1
macOS macOS 26.4
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,501 96.7 20.4 GB view
4k 1,579 95.4 21.2 GB view
8k 1,919 88.8 21.7 GB view
16k 1,884 84.2 22.4 GB current
Batching Results
Batch Size TG tok/s Speedup
96.7 1.00×
142.0 1.47×
185.6 1.92×