Microsoft
phi-3-medium 14B benchmark results
5 results on 4 benchmarks from 2 sources.
Open weights Released: Apr 23, 2024 Updated Oct 22, 2026
Also listed as: Phi-3-medium-128k-instruct · Phi-3 (14B) · Phi-3 14b · Phi-3-medium · microsoft/Phi-3-medium-128k-instruct
All benchmark results
| Benchmarks | Score | Rank | Setting | Source type | Date |
|---|---|---|---|---|---|
| Open LLM Leaderboard v2 | 32.0% | 847 | Average of 6 normalised benchmarks; bfloat16; 13.96B params“microsoft/Phi-3-medium-128k-instruct” | Official leaderboard | Mar 13, 2025 |
| GPQA Diamond | 27.6% | 266 | Epoch AI run, accuracy | Independent run | Jan 31, 2025 |
| MMLU | 77.9% | 15 | Exact match; 5-shot | Paper / tech report via Epoch AI | Apr 23, 2024 |
| MMLU | 77.5% | 31 | Exact match; 5-shot | Official leaderboard via Epoch AI | Apr 23, 2024 |
| MATH Level 5 | 17.6% | 89 | Epoch AI run, accuracy | Independent run | Jan 31, 2025 |
FAQ
Where do these phi-3-medium 14B scores come from?
From Epoch AI Benchmarking Hub and Hugging Face Open LLM Leaderboard v2 (archived). Each row links to the original leaderboard, dataset or report, and vendor-reported numbers are labelled.