Phi-3 Medium Instruct 14B

Benchmark scores, pricing, speed and comparisons for Phi-3 Medium Instruct 14B.

Phi-3 Medium Instruct 14B is an AI model from Microsoft Azure, released 2024-05-21. It scores 14.4 on the Intelligence Index, 8.9 on the Coding Index and 1.3 on the Math Index.

Phi-3 Medium Instruct 14B at a glance

Microsoft Azure released Phi-3 Medium Instruct 14B on May 21, 2024. It is marked as retired in the catalogue, so treat its numbers as a historical record.

Phi-3 Medium Instruct 14B scores 14.4 on the Intelligence Index, the overall score the AI Benchmark rankings sort models by. On the narrower indexes it scores 8.9 for coding and 1.3 for math. Of its 12 individual benchmark results, the highest is 54.3% on MMLU Pro and the lowest is 0.0% on TAU2. These scores come from independent third-party model evaluations, which Franklin AI syncs automatically each day.

It costs $0.17 per million input tokens and $0.68 per million output tokens, so a sample workload of 1 million input tokens and 200,000 output tokens costs about $0.31.

The benchmark data has no speed measurement for Phi-3 Medium Instruct 14B yet.

Every pair of models in the catalogue has a head-to-head page. Pick Phi-3 Medium Instruct 14B and a rival on the model comparisons page to compare their scores, prices and speed side by side.

Key scores

Intelligence14.4Index score
Coding8.9Index score
Math1.3Index score
Blended Cost$0.30Per 1M tokens

How does Phi-3 Medium Instruct 14B score on each benchmark?

12 benchmark results for Phi-3 Medium Instruct 14B.

MMLU Pro Knowledge
54.3%
MATH 500 Math
46.3%
GPQA Reasoning
32.6%
IFBench Other
21.5%
LiveCodeBench Code
15.0%
SciCode Other
11.8%
HLE Other
4.5%
LCR Other
3.0%
AIME Math
1.3%
AIME 2025 Math
1.3%
TerminalBench Hard Other
0.0%
TAU2 Other
0.0%