← Back to community benchmarks

Qwopus3.5-9B-v3

M1 Pro (14c) · 16 GB · 4bit · 2026-05-04
Performance
8k
tokens
138.8
PP tok/s
28.4
TG tok/s
59007
TTFT (ms)
6.9
Peak mem (GB)
Hardware
Chip M1 Pro (14c)
Memory 16 GB
GPU Cores 14
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 134.1 30.6 5.8 GB view
4k 139.3 29.5 6.5 GB view
8k 138.8 28.4 6.9 GB current
16k 136.5 26.2 7.9 GB view
32k 131.0 22.8 9.7 GB view