# Phi-4-reasoning benchmark scores

Other · model · https://benchgap.net/model/phi-4-reasoning · data retrieved 2026-10-29

As of 2026-10-29, Phi-4-reasoning (Other) has measured scores on 12 benchmarks and estimated scores on 32 more.

| Benchmark | Score | Source |
|---|---|---|
| [AA-GPQA Diamond](https://benchgap.net/b/aa-gpqa-diamond/current.md) | 57.5% | measured |
| [AA-HLE](https://benchgap.net/b/aa-hle/current.md) | 3.8% | measured |
| [AIME26](https://benchgap.net/b/aime2026/current.md) | 64.9% | estimated ± 4.6 pp, medium confidence |
| [CritPt](https://benchgap.net/b/critpt/current.md) | 0.0% | measured |
| [GDPval-AA](https://benchgap.net/b/gdpvalaanormalized/current.md) | 0.0% | estimated ± 10.9 pp, low confidence |
| [GPQA Diamond](https://benchgap.net/b/gpqa-diamond/current.md) | 66.1% | estimated ± 1.2 pp, high confidence |
| [HLE](https://benchgap.net/b/hle/current.md) | 6.9% | estimated ± 6.9 pp, medium confidence |
| [IFBench](https://benchgap.net/b/ifbench/current.md) | 53.6% | estimated ± 6.6 pp, low confidence |
| [AA-LCR](https://benchgap.net/b/lcr/current.md) | 0.0% | measured |
| [IFEval](https://benchgap.net/b/ifeval/current.md) | 83.4% | measured |
| [LongBench v2](https://benchgap.net/b/longbench-v2/current.md) | 47.1% | estimated ± 5.8 pp, low confidence |
| [τ²-bench results](https://benchgap.net/b/tau2-bench/current.md) | 0.0% | measured |
| [BioMysteryBench (human-difficult)](https://benchgap.net/b/biomysterybenchhumandifficult/current.md) | 0.0% | estimated ± 6.8 pp, low confidence |
| [SWE-bench Verified](https://benchgap.net/b/swe-bench-verified/current.md) | 54.7% | estimated ± 9.3 pp, low confidence |
| [AA Agentic Index](https://benchgap.net/b/aa-agentic-index/current.md) | 0.0% | estimated ± 12.3 pp, low confidence |
| [MMLU-Pro](https://benchgap.net/b/mmlu-pro/current.md) | 74.3% | measured |
| [AA-IFBench](https://benchgap.net/b/aa-if-bench/current.md) | 23.5% | measured |
| [AIME 2024](https://benchgap.net/b/aime2024/current.md) | 75.3% | measured |
| [Artificial Analysis Intelligence Index](https://benchgap.net/b/artificialanalysis/current.md) | 10.3% | estimated ± 5.8 pp, medium confidence |
| [BBH](https://benchgap.net/b/bbh/current.md) | 92.4% | estimated ± 2.8 pp, medium confidence |
| [GPQA](https://benchgap.net/b/gpqa/current.md) | 65.8% | measured |
| [MMLU](https://benchgap.net/b/mmlu/current.md) | 86.1% | estimated ± 2.2 pp, high confidence |
| [AA-Omniscience Accuracy](https://benchgap.net/b/omniscienceaccuracy/current.md) | 20.9% | estimated ± 7.8 pp, medium confidence |
| [MATH-500](https://benchgap.net/b/math-500/current.md) | 96.8% | estimated ± 1.2 pp, high confidence |
| [MMMLU](https://benchgap.net/b/mmmlu/current.md) | 83.3% | estimated ± 0.7 pp, medium confidence |
| [AA MMLU-Pro](https://benchgap.net/b/aa-mmlu-pro/current.md) | 84.3% | estimated ± 1.0 pp, low confidence |
| [AIME 2025](https://benchgap.net/b/aime2025/current.md) | 62.9% | measured |
| [ARC-AGI-1](https://benchgap.net/b/arcagi1/current.md) | 74.7% | estimated ± 7.8 pp, low confidence |
| [ARC-AGI-2](https://benchgap.net/b/arc-agi-2/current.md) | 30.7% | estimated ± 14.6 pp, low confidence |
| [Vals GPQA Diamond](https://benchgap.net/b/vals-gpqa-diamond/current.md) | 43.8% | estimated ± 7.1 pp, low confidence |
| [Vals MMLU-Pro](https://benchgap.net/b/vals-mmlu-pro/current.md) | 80.1% | estimated ± 2.3 pp, low confidence |
| [HLE w/o tools](https://benchgap.net/b/hlenotools/current.md) | 11.6% | estimated ± 5.2 pp, low confidence |
| [ARC-AGI-3](https://benchgap.net/b/arcagi3/current.md) | 1.4% | estimated ± 7.7 pp, low confidence |
| [C-Eval](https://benchgap.net/b/ceval/current.md) | 90.8% | estimated ± 1.1 pp, low confidence |
| [HMMT Feb 2025](https://benchgap.net/b/hmmtfeb2025/current.md) | 29.9% | estimated ± 3.1 pp, medium confidence |
| [MMLU-Redux](https://benchgap.net/b/mmluredux/current.md) | 86.9% | estimated ± 0.6 pp, high confidence |
| [SuperGPQA](https://benchgap.net/b/supergpqa/current.md) | 46.3% | estimated ± 8.2 pp, low confidence |
| [HealthBench Hard](https://benchgap.net/b/healthbench-hard/current.md) | 0.0% | estimated ± 8.6 pp, low confidence |
| [MMLU-Pro (Arcee)](https://benchgap.net/b/mmlu-pro-arcee/current.md) | 77.1% | estimated ± 4.0 pp, low confidence |
| [BioMysteryBench (human-solvable)](https://benchgap.net/b/biomysterybenchhumansolvable/current.md) | 81.7% | estimated ± 0.6 pp, low confidence |
| [STEMBench](https://benchgap.net/b/stembench/current.md) | 2.4% | estimated ± 0.2 pp, medium confidence |
| [SimpleQA](https://benchgap.net/b/simpleqa/current.md) | 17.2% | estimated ± 7.8 pp, medium confidence |
| [LiveCodeBench](https://benchgap.net/b/livecodebench/current.md) | 53.8% | measured |
| [MRCRv2](https://benchgap.net/b/mrcrv2/current.md) | 22.8% | estimated ± 11.3 pp, low confidence |