← Back to community benchmarks

Qwen3.5-4B-OptiQ

M3 (10c) · 16 GB · 4bit · 2026-05-18
Performance
4k
tokens
331.7
PP tok/s
24.8
TG tok/s
12349
TTFT (ms)
4.7
Peak mem (GB)
Hardware
Chip M3 (10c)
Memory 16 GB
GPU Cores 10
Software
oMLX v0.3.9.dev1
macOS macOS 26.2
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 319.0 25.6 4.1 GB view
4k 331.7 24.8 4.7 GB current
Batching Results
Batch Size TG tok/s Speedup
25.6 1.00×
48.5 1.89×
66.7 2.61×