← Back to community benchmarks

Qwen3.5-35B-A3B

M5 Pro (20c) · 64 GB · 8bit · 2026-03-26
Performance
4k
tokens
2,151
PP tok/s
47.5
TG tok/s
1904
TTFT (ms)
37.1
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 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,535 49.5 36.2 GB view
4k 2,151 47.5 37.1 GB current
8k 2,168 44.7 37.5 GB view
16k 2,031 40.2 38.2 GB view
32k 1,727 31.9 39.5 GB view
64k 1,253 23.8 42.1 GB view
128k 801.7 15.5 47.5 GB view