← Back to community benchmarks

Qwen3.5-35B-A3B

M5 Pro (20c) · 64 GB · 8bit · 2026-03-26
Performance
128k
tokens
802.2
PP tok/s
22.9
TG tok/s
163393
TTFT (ms)
47.5
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.2.21
macOS macOS 26.3.1
Context 131,072
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,844 51.6 36.2 GB view
4k 2,219 52.0 37.1 GB view
8k 2,193 50.8 37.5 GB view
16k 2,002 47.8 38.2 GB view
32k 1,732 40.8 39.5 GB view
64k 1,253 31.9 42.1 GB view
128k 802.2 22.9 47.5 GB current