← Back to community benchmarks

Qwen3-14B

M4 (10c) · 32 GB · 8bit · 2026-05-12
Performance
4k
tokens
113.8
PP tok/s
6.6
TG tok/s
35991
TTFT (ms)
15.6
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 32 GB
GPU Cores 10
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 116.8 6.9 14.8 GB view
4k 113.8 6.6 15.6 GB current
Batching Results
Batch Size TG tok/s Speedup
6.9 1.00×
13.7 1.99×
20.8 3.01×