Phi-4-reasoning places in the top three on HumanEval+.
Microsoft Apr 2025 Open weights Small
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Arena-Hard-Auto | 73.3% | #24 of 39 | Model card | |
| MMLU-Pro | 74.3% | #103 of 151 | Model card | |
| IFEval | 83.4% | #86 of 126 | Model card | |
| FlenQA | 97.7% | #2 of 2 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| 3SAT-Search | 78% | #1 of 2 | Model card | |
| BA-Calendar | 67.7% | #1 of 2 | Model card | |
| AIME 2024 | 75.3% | #52 of 87 | Model card | |
| GPQA Diamond | 67.1% | #182 of 284 | Model card | |
| Omni-MATH | 76.6% | #3 of 4 | Model card | |
| AIME 2025 | 63.1% | #122 of 143 | Model card | |
| HMMT 2025 | 43.8% | #48 of 49 | Model card | |
| AIME 1983-2024 | 83.1% | #2 of 2 | Model card | |
| TSP-Opt | 37.5% | #2 of 2 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| HumanEval+ | 92.9% | #3 of 27 | Model card | |
| LiveCodeBench | 53.8% | #38 of 73 | Model card | |
| Codeforces rating | 1,736 | #31 of 41 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SpatialEval Maze-Nav | 55.1% | #1 of 2 | Model card | |
| SpatialEval Spatial-Map | 73.7% | #1 of 2 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| ToxiGen (discriminative) | 85.5% | #1 of 2 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| PhiBench | 70.6% | #2 of 3 | Model card |
The benchmarks the lab chose to headline, with the scores it gave.
Every result in the longer documents the lab published with it. 9 of them aren't in the launch post.