← Back to community benchmarks

Qwen3.6-35B-A3B

M5 Pro (20c) · 48 GB · 4bit · 2026-04-17
Performance
4k
tokens
1,929
PP tok/s
84.4
TG tok/s
2124
TTFT (ms)
20.0
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 48 GB
GPU Cores 20
Software
oMLX v0.3.5
macOS macOS 26.5
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,312 89.9 19.2 GB view
4k 1,929 84.4 20.0 GB current
8k 2,058 81.1 20.4 GB view
16k 1,959 71.7 21.0 GB view
Batching Results
Batch Size TG tok/s Speedup
89.9 1.00×
141.1 1.57×
279.6 3.11×