← Back to community benchmarks

Qwen3.5-4B

M2 (10c) · 24 GB · 4bit · 2026-04-18
Performance
1k
tokens
191.6
PP tok/s
31.3
TG tok/s
5345
TTFT (ms)
3.3
Peak mem (GB)
Hardware
Chip M2 (10c)
Memory 24 GB
GPU Cores 10
Software
oMLX v0.3.6
macOS macOS 15.7.5
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 191.6 31.3 3.3 GB current
4k 196.2 30.1 3.9 GB view
8k 193.6 28.0 4.3 GB view
Batching Results
Batch Size TG tok/s Speedup
31.3 1.00×
44.5 1.42×
46.4 1.48×
39.8 1.27×