← Back to community benchmarks

Qwen3.5-4B

M1 (8c) · 16 GB · 4bit · 2026-03-09
Performance
32k
tokens
123.6
PP tok/s
14.0
TG tok/s
265026
TTFT (ms)
5.9
Peak mem (GB)
Hardware
Chip M1 (8c)
Memory 16 GB
GPU Cores 8
Software
oMLX v0.2.6
macOS macOS 26.3
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 146.0 22.7 3.9 GB view
4k 144.9 21.5 4.2 GB view
16k 135.7 17.1 4.8 GB view
32k 123.6 14.0 5.9 GB current
Batching Results
Batch Size TG tok/s Speedup
22.7 1.00×
33.3 1.47×
34.5 1.52×