← Back to community benchmarks

Qwen3.5-4B

M4 Pro (20c) · 48 GB · 4bit · 2026-03-09
Performance
32k
tokens
655.0
PP tok/s
49.9
TG tok/s
50030
TTFT (ms)
6.9
Peak mem (GB)
Hardware
Chip M4 Pro (20c)
Memory 48 GB
GPU Cores 20
Software
oMLX v0.2.6
macOS macOS 26.3.1
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 811.1 82.1 4.0 GB view
4k 798.0 76.8 4.2 GB view
8k 781.9 72.4 4.6 GB view
16k 737.3 62.8 5.3 GB view
32k 655.0 49.9 6.9 GB current
64k 534.5 36.3 8.6 GB view
Batching Results
Batch Size TG tok/s Speedup
82.1 1.00×
131.7 1.60×
172.4 2.10×