← Back to community benchmarks

Qwen3.5-4B

M4 (10c) · 24 GB · 4bit · 2026-03-09
Performance
64k
tokens
247.3
PP tok/s
14.5
TG tok/s
264997
TTFT (ms)
8.0
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 24 GB
GPU Cores 10
Software
oMLX v0.2.6
macOS macOS 15.6
Context 65,536
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 382.7 36.6 3.9 GB view
16k 340.3 27.7 4.8 GB view
64k 247.3 14.5 8.0 GB current
Batching Results
Batch Size TG tok/s Speedup
36.6 1.00×
71.8 1.96×
89.1 2.43×
93.1 2.54×