← Back to community benchmarks

Qwen3.5-4B

M5 (8c) · 16 GB · 4bit · 2026-04-18
Performance
1k
tokens
902.3
PP tok/s
48.1
TG tok/s
1135
TTFT (ms)
3.3
Peak mem (GB)
Hardware
Chip M5 (8c)
Memory 16 GB
GPU Cores 8
Software
oMLX v0.3.6
macOS macOS 26.4.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 902.3 48.1 3.3 GB current
4k 994.2 46.3 3.9 GB view
Batching Results
Batch Size TG tok/s Speedup
48.1 1.00×
85.9 1.79×