← Back to community benchmarks

Qwen3.5-4B

M3 (10c) · 16 GB · 4bit · 2026-03-10
Performance
1k
tokens
309.5
PP tok/s
33.4
TG tok/s
3309
TTFT (ms)
3.3
Peak mem (GB)
Hardware
Chip M3 (10c)
Memory 16 GB
GPU Cores 10
Software
oMLX v0.2.8b1
macOS macOS 26.3
Context 1,024
Batching Results
Batch Size TG tok/s Speedup
33.4 1.00×
65.1 1.95×
74.0 2.22×