ModelDen

Phi-4 14B

Microsoft · 14.7B · Q4_K_M · ~8.9 GB
Released December 12, 2024

How good is it? — per-capability quality

Each number is the model’s score on its best public benchmarkfor that skill — named and linked below, never a blend. Scores compare cleanly when the benchmark matches; benchmarks are gameable, so we show the receipts rather than hide behind one number.

Coding83
HumanEval+ · source · as of 2026-06
Math & logic80
MATH · source · as of 2026-06
General56
GPQA · source · as of 2026-06

Microsoft's 14B model punching well above its size on reasoning and math

About this model

Developer
Microsoft
Family
Phi-4
License
MIT
Context window
16,384 tokens
Will Phi-4 14B actually run on YOUR graphics card?

This page shows fit for a generic VRAM tier. Tell us your GPU for a personalized speed estimate and fit verdict.

Enter your graphics card →