← Back to community benchmarks

Qwen3.5-4B

M2 (10c) · 16 GB · 4bit · 2026-03-10
Performance
4k
tokens
199.8
PP tok/s
29.6
TG tok/s
20504
TTFT (ms)
4.2
Peak mem (GB)
Hardware
Chip M2 (10c)
Memory 16 GB
GPU Cores 10
Software
oMLX v0.2.6
macOS macOS 26.3.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 199.8 31.2 3.9 GB view
4k 199.8 29.6 4.2 GB current
Batching Results
Batch Size TG tok/s Speedup
31.2 1.00×
44.3 1.42×
46.8 1.50×