← Back to community benchmarks

Qwen3.5-4B

M4 (10c) · 16 GB · 4bit · 2026-03-09
Performance
8k
tokens
385.2
PP tok/s
31.8
TG tok/s
21268
TTFT (ms)
4.6
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 16 GB
GPU Cores 10
Software
oMLX v0.2.6
macOS macOS 15.7.3
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 406.1 37.4 3.9 GB view
4k 403.8 36.0 4.2 GB view
8k 385.2 31.8 4.6 GB current
Batching Results
Batch Size TG tok/s Speedup
37.4 1.00×
70.9 1.90×
95.0 2.54×