← Back to community benchmarks

Qwen3.5-4B

M4 · 14 GB · 4bit · 2026-03-10
Performance
4k
tokens
365.9
PP tok/s
35.8
TG tok/s
11196
TTFT (ms)
4.2
Peak mem (GB)
Hardware
Chip M4
Memory 14 GB
Software
oMLX v0.2.7
macOS macOS 26.2
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 377.0 38.0 3.9 GB view
4k 365.9 35.8 4.2 GB current
Batching Results
Batch Size TG tok/s Speedup
38.0 1.00×
62.0 1.63×
89.6 2.36×