← Back to community benchmarks

Qwen3.5-4Blm

M3 Pro (14c) · 18 GB · 4bit · 2026-03-10
Performance
4k
tokens
459.5
PP tok/s
46.7
TG tok/s
8913
TTFT (ms)
3.5
Peak mem (GB)
Hardware
Chip M3 Pro (14c)
Memory 18 GB
GPU Cores 14
Software
oMLX v0.2.7
macOS macOS 26.3.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 438.6 49.8 3.3 GB view
4k 459.5 46.7 3.5 GB current
Batching Results
Batch Size TG tok/s Speedup
49.8 1.00×
91.3 1.83×
119.0 2.39×