← Back to community benchmarks

Qwen2.5-Coder-14B-Instruct

M4 (10c) · 32 GB · 4bit · 2026-05-18
Performance
4k
tokens
119.7
PP tok/s
11.4
TG tok/s
34226
TTFT (ms)
9.2
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 32 GB
GPU Cores 10
Software
oMLX v0.3.9.dev2
macOS macOS 26.4.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 122.7 12.0 8.4 GB view
4k 119.7 11.4 9.2 GB current
Batching Results
Batch Size TG tok/s Speedup
12.0 1.00×
24.3 2.02×
29.2 2.43×