← Back to community benchmarks

Qwen3.5-35B-A3B

M5 Pro (20c) · 64 GB · 4bit · 2026-03-29
Performance
4k
tokens
2,328
PP tok/s
71.7
TG tok/s
1760
TTFT (ms)
21.1
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.2.24
macOS macOS 26.4
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,992 74.7 20.1 GB view
4k 2,328 71.7 21.1 GB current
8k 2,315 68.2 21.4 GB view
16k 2,127 62.3 22.0 GB view
32k 1,791 52.8 23.4 GB view
Batching Results
Batch Size TG tok/s Speedup
74.7 1.00×
120.0 1.61×
168.6 2.26×