← Back to community benchmarks

Qwen3.5-4B-Uncensored

M5 (10c) · 24 GB · 4bit · 2026-04-15
Performance
32k
tokens
805.6
PP tok/s
30.6
TG tok/s
40675
TTFT (ms)
6.5
Peak mem (GB)
Hardware
Chip M5 (10c)
Memory 24 GB
GPU Cores 10
Software
oMLX v0.3.5
macOS macOS 26.4
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
4k 1,130 45.5 3.9 GB view
8k 1,100 43.1 4.3 GB view
16k 1,004 38.3 5.0 GB view
32k 805.6 30.6 6.5 GB current