Other · model
Phi-4-reasoning-plus benchmark scores
As of 2026-10-29, Phi-4-reasoning-plus (Other) has measured scores on 6 benchmarks and estimated scores on 22 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 66.8% | estimated ± 6.3 pp, medium confidence |
| AA-HLE | 6.9% | estimated ± 5.0 pp, high confidence |
| AIME26 | 78.8% | estimated ± 4.6 pp, medium confidence |
| CritPt | 0.0% | estimated ± 3.4 pp, medium confidence |
| GPQA Diamond | 69.1% | estimated ± 1.2 pp, high confidence |
| HLE | 8.4% | estimated ± 6.9 pp, medium confidence |
| IFBench | 54.7% | estimated ± 9.2 pp, medium confidence |
| IFEval | 84.9% | measured |
| SWE-bench Verified | 54.4% | estimated ± 9.3 pp, low confidence |
| MMLU-Pro | 76.0% | measured |
| AA-IFBench | 33.6% | estimated ± 8.8 pp, medium confidence |
| AIME 2024 | 81.3% | measured |
| Artificial Analysis Intelligence Index | 11.0% | estimated ± 6.3 pp, medium confidence |
| BBH | 93.5% | estimated ± 2.8 pp, medium confidence |
| GPQA | 68.9% | measured |
| MMLU | 86.9% | estimated ± 2.2 pp, high confidence |
| AA-Omniscience Accuracy | 21.7% | estimated ± 8.1 pp, medium confidence |
| MATH-500 | 97.3% | estimated ± 1.2 pp, high confidence |
| MMMLU | 82.5% | estimated ± 1.7 pp, high confidence |
| AA MMLU-Pro | 84.7% | estimated ± 1.0 pp, low confidence |
| AIME 2025 | 78.0% | measured |
| HLE w/o tools | 11.7% | estimated ± 5.2 pp, low confidence |
| C-Eval | 88.2% | estimated ± 1.2 pp, high confidence |
| HMMT Feb 2025 | 65.5% | estimated ± 3.1 pp, medium confidence |
| MMLU-Redux | 88.0% | estimated ± 0.6 pp, high confidence |
| SuperGPQA | 49.0% | estimated ± 8.2 pp, low confidence |
| SimpleQA | 27.6% | estimated ± 13.1 pp, low confidence |
| LiveCodeBench | 53.1% | measured |