← Back to community benchmarks

Qwen3.6-35B-A3B-UD

M3 Pro (18c) · 36 GB · 4bit · 2026-05-15
Performance
1k
tokens
574.4
PP tok/s
37.0
TG tok/s
1783
TTFT (ms)
20.4
Peak mem (GB)
Hardware
Chip M3 Pro (18c)
Memory 36 GB
GPU Cores 18
Software
oMLX v0.3.8
macOS macOS 15.7.5
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 574.4 37.0 20.4 GB current
4k 659.4 36.2 21.2 GB view
Batching Results
Batch Size TG tok/s Speedup
37.0 1.00×
60.8 1.64×
88.4 2.39×