← Back to community benchmarks

Qwen3.5-35B-A3B

M5 Pro (20c) · 64 GB · 8bit · 2026-03-27
Performance
16k
tokens
2,053
PP tok/s
40.9
TG tok/s
7979
TTFT (ms)
38.2
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.2.22
macOS macOS 26.3.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,769 50.9 36.2 GB view
4k 2,224 48.2 37.1 GB view
8k 2,218 45.6 37.5 GB view
16k 2,053 40.9 38.2 GB current
32k 1,748 31.5 39.5 GB view
64k 1,254 23.0 42.1 GB view
128k 804.4 14.6 47.5 GB view