OpenAI · model
GPT-5.4 mini benchmark scores
As of 2026-11-22, GPT-5.4 mini (OpenAI) has measured scores on 41 benchmarks and estimated scores on 140 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 87.5% | measured |
| AA-HLE | 28.1% | measured |
| AA-SciCode | 52.1% | measured |
| AIME26 | 95.9% | estimated ± 1.2 pp, medium confidence |
| Artificial Analysis Intelligence Index | 24.1% | measured |
| CritPt | 10.0% | measured |
| GDPval-AA | 25.6% | measured |
| GPQA Diamond | 88.0% | estimated ± 0.1 pp, medium confidence |
| HLE | 41.5% | measured |
| IFBench | 71.3% | estimated ± 6.6 pp, medium confidence |
| AA-LCR | 77.0% | measured |
| LiveCodeBench v6 | 87.1% | estimated ± 2.2 pp, medium confidence |
| AA-Omniscience Accuracy | 37.5% | measured |
| SciCode | 39.7% | estimated ± 2.9 pp, low confidence |
| Terminal-Bench 2.1 | 70.6% | estimated ± 5.6 pp, medium confidence |
| BrowseComp | 82.3% | estimated ± 4.2 pp, high confidence |
| HLE w/ tools | 56.8% | estimated ± 4.1 pp, medium confidence |
| IFEval | 92.9% | estimated ± 2.8 pp, high confidence |
| LongBench v2 | 63.2% | estimated ± 5.5 pp, medium confidence |
| VITA-Bench | 38.5% | estimated ± 9.9 pp, low confidence |
| τ²-bench results | 93.4% | measured |
| AA Coding Index | 56.1% | measured |
| AA-MMMU-Pro | 73.3% | measured |
| APEX-Agents | 30.6% | estimated ± 4.9 pp, low confidence |
| APEX-Agents-AA | 28.2% | measured |
| BioMysteryBench (human-difficult) | 43.8% | estimated ± 6.8 pp, low confidence |
| DeepSearchQA | 84.0% | estimated ± 4.5 pp, high confidence |
| SWE-bench Verified | 80.5% | estimated ± 2.2 pp, medium confidence |
| BFCL v4 | 73.5% | estimated ± 2.7 pp, high confidence |
| CyberGym | 60.6% | estimated ± 5.3 pp, low confidence |
| JobBench | 32.0% | estimated ± 7.3 pp, low confidence |
| SkillsBench | 45.6% | estimated ± 9.1 pp, low confidence |
| SWE-bench Pro | 56.8% | estimated ± 2.9 pp, medium confidence |
| WideResearch | 81.4% | estimated ± 2.1 pp, medium confidence |
| AA Agentic Index | 19.7% | measured |
| MMLU-Pro | 85.1% | estimated ± 2.5 pp, medium confidence |
| AA-IFBench | 73.3% | measured |
| FailureSensorIQ | 66.0% | estimated ± 1.7 pp, high confidence |
| GPQA | 88.0% | measured |
| MMLU | 90.5% | estimated ± 0.9 pp, medium confidence |
| Vals GPQA Diamond | 83.1% | measured |
| Vals LiveCodeBench | 81.5% | measured |
| Vals MMLU-Pro | 84.6% | measured |
| MMMLU | 89.5% | estimated ± 0.7 pp, medium confidence |
| MMMU | 82.8% | estimated ± 0.8 pp, high confidence |
| Chess Puzzles | 17.0% | measured |
| AA MMLU-Pro | 87.0% | estimated ± 0.7 pp, high confidence |
| Gert Labs | 47.7% | estimated ± 9.3 pp, medium confidence |
| AA Harvey LAB v1.0 | 80.8% | estimated ± 1.2 pp, low confidence |
| AA Tau3 Banking | 39.6% | estimated ± 2.0 pp, low confidence |
| AA Terminal-Bench 2.1 | 80.0% | estimated ± 2.5 pp, medium confidence |
| ApprenticeBench | 3.5% | estimated ± 5.0 pp, low confidence |
| ARC-AGI-1 | 63.7% | measured |
| ARC-AGI-2 | 18.9% | measured |
| CursorBench 3.1 | 49.2% | estimated ± 3.4 pp, medium confidence |
| CursorBench 3.2 | 41.0% | estimated ± 4.4 pp, medium confidence |
| EBR-bench | 4.2% | estimated ± 8.1 pp, low confidence |
| Terminal-Bench 3.0 | 8.3% | estimated ± 3.2 pp, medium confidence |
| FrontierCode 1.1 Main | 27.0% | measured |
| FrontierMath v2 (Tier 4) | 9.8% | measured |
| FrontierMath v2 (Tiers 1-3) | 51.2% | measured |
| FrontierSWE v2 | 7.3% | estimated ± 9.5 pp, low confidence |
| Furniture Assembly | 32.9% | estimated ± 9.3 pp, medium confidence |
| MaxxBench | 47.2% | estimated ± 8.8 pp, medium confidence |
| MazeBench | 0.4% | estimated ± 5.5 pp, low confidence |
| Mystery Game Puzzles | 7.0% | measured |
| NYT Connections | 61.8% | measured |
| OfficeQA Pro | 50.3% | estimated ± 6.8 pp, low confidence |
| OSWorld-Verified | 72.1% | measured |
| ParseBench-1 | 59.0% | estimated ± 5.5 pp, low confidence |
| Terminal-Bench 4.1 | 1.8% | measured |
| Terminal-Bench Hard | 52.3% | measured |
| Vals SWE-bench | 73.0% | measured |
| Terminal-Bench 2.1 (Vals AI) | 54.7% | measured |
| VulcanBench v3 | 81.6% | estimated ± 5.6 pp, low confidence |
| CWE-Bench-AA | 33.1% | estimated ± 10.1 pp, low confidence |
| AA Cyber Index | 13.3% | estimated ± 9.2 pp, low confidence |
| DeepsecBench-AA | 1.5% | estimated ± 8.9 pp, low confidence |
| HealthBench Professional (length-adjusted, Artificial Analysis) | 53.1% | estimated ± 4.3 pp, low confidence |
| HealthBench Professional (raw, Artificial Analysis) | 55.1% | estimated ± 0.8 pp, low confidence |
| AA ITBench | 35.2% | measured |
| Bug Hunt Bench | 5.8% | estimated ± 9.0 pp, low confidence |
| Chartography (tools) | 79.2% | estimated ± 4.1 pp, low confidence |
| CursorBench 4.0 | 30.8% | estimated ± 3.4 pp, medium confidence |
| CWE-bench v1 | 35.1% | estimated ± 6.4 pp, low confidence |
| DeepSWE | 24.7% | estimated ± 6.4 pp, low confidence |
| DRACO | 73.0% | estimated ± 2.0 pp, medium confidence |
| FrontierCode 2.0 | 26.1% | estimated ± 9.3 pp, low confidence |
| FrontierPhysics | 2.5% | estimated ± 5.0 pp, low confidence |
| HLE w/o tools | 28.2% | measured |
| OSWorld 2.0 | 6.2% | estimated ± 3.6 pp, high confidence |
| OSWorld 2.1 Offline | 2.8% | estimated ± 1.9 pp, low confidence |
| PostTrainBench v1.1 | 18.0% | estimated ± 3.7 pp, medium confidence |
| SpireClimb | 5.8% | estimated ± 7.2 pp, low confidence |
| StealthMark | 32.1% | estimated ± 9.2 pp, medium confidence |
| Terminal-Bench 4.0 | 10.7% | estimated ± 4.5 pp, medium confidence |
| Toolathlon-Verified | 47.7% | estimated ± 2.3 pp, low confidence |
| VulcanBench Frontier v4 | 82.2% | estimated ± 6.8 pp, low confidence |
| GDP.pdf | 17.0% | estimated ± 4.6 pp, medium confidence |
| AA Terminal-Bench 4.0 | 1.4% | estimated ± 6.2 pp, low confidence |
| ARC-AGI-3 | 0.4% | estimated ± 7.2 pp, low confidence |
| ARC-AGI-3 Provider Adapter | 2.7% | estimated ± 7.3 pp, low confidence |
| cursorBench41 | 44.8% | estimated ± 3.7 pp, medium confidence |
| MKA | 21.7% | estimated ± 6.4 pp, low confidence |
| SkillsBench official leaderboard (BenchFlow) | 41.4% | measured |
| BinaryAudit | 25.6% | estimated ± 12.0 pp, low confidence |
| CompileBench | 55.6% | estimated ± 10.0 pp, low confidence |
| OTelBench | 22.5% | estimated ± 6.6 pp, medium confidence |
| Vibe Code Bench | 48.0% | measured |
| FACTS Multimodal | 35.5% | estimated ± 4.0 pp, high confidence |
| Hermes Bench | 56.3% | estimated ± 2.9 pp, high confidence |
| Hermes Index | 23.4% | estimated ± 8.7 pp, medium confidence |
| ParseBench-2 | 67.1% | estimated ± 5.0 pp, medium confidence |
| CharXiv | 78.2% | estimated ± 5.6 pp, medium confidence |
| CharXiv w/o tools | 80.7% | estimated ± 1.2 pp, high confidence |
| ExploitBench | 14.6% | estimated ± 7.2 pp, low confidence |
| ExploitGym | 1.0% | estimated ± 10.6 pp, low confidence |
| AI2D_TEST | 89.3% | estimated ± 2.0 pp, medium confidence |
| C-Eval | 92.8% | estimated ± 1.0 pp, medium confidence |
| Claw-Eval | 64.5% | estimated ± 6.2 pp, low confidence |
| CountBench | 93.8% | estimated ± 2.9 pp, medium confidence |
| DeepPlanning | 38.8% | estimated ± 13.1 pp, low confidence |
| DynaMath | 84.5% | estimated ± 3.0 pp, medium confidence |
| ERQA | 55.9% | estimated ± 3.1 pp, high confidence |
| HMMT Feb 2026 | 87.2% | estimated ± 1.5 pp, medium confidence |
| HMMT Nov 2025 | 94.5% | estimated ± 1.5 pp, medium confidence |
| MathVision | 85.8% | estimated ± 3.0 pp, high confidence |
| MCP Atlas | 57.7% | measured |
| MCP-Tasks | 76.2% | estimated ± 0.6 pp, low confidence |
| MLVU (M-Avg) | 85.3% | estimated ± 2.2 pp, high confidence |
| MMAnswerBench | 86.9% | estimated ± 2.4 pp, medium confidence |
| MMLU-Redux | 95.7% | estimated ± 0.8 pp, medium confidence |
| MMMU-Pro | 76.6% | measured |
| NL2Repo | 41.5% | estimated ± 4.7 pp, high confidence |
| PostTrainBench v1 | 19.6% | estimated ± 3.1 pp, medium confidence |
| QwenClawBench | 55.2% | estimated ± 2.6 pp, high confidence |
| RealWorldQA | 83.4% | estimated ± 2.8 pp, high confidence |
| ScreenSpot Pro | 74.5% | estimated ± 9.6 pp, low confidence |
| SimpleVQA | 64.6% | estimated ± 7.5 pp, medium confidence |
| SuperGPQA | 81.3% | estimated ± 7.2 pp, medium confidence |
| Terminal-Bench 2.0 | 60.0% | measured |
| Toolathlon | 42.9% | measured |
| V* | 100.0% | estimated ± 7.5 pp, medium confidence |
| Video-MME (w/o subtitle) | 83.7% | estimated ± 1.3 pp, high confidence |
| Video-MME (with subtitle) | 86.2% | estimated ± 3.0 pp, high confidence |
| VideoMMMU | 84.5% | estimated ± 2.4 pp, high confidence |
| We-Math | 81.6% | estimated ± 7.4 pp, low confidence |
| AA LiveCodeBench | 88.0% | estimated ± 2.6 pp, high confidence |
| HealthBench Hard | 0.0% | estimated ± 5.1 pp, low confidence |
| MedXpertQA (MM) | 67.0% | estimated ± 2.4 pp, high confidence |
| MedXpertQA (Text) | 44.3% | estimated ± 8.9 pp, low confidence |
| React Native Evals | 81.1% | estimated ± 3.0 pp, medium confidence |
| ResearchClawBench | 17.1% | estimated ± 1.8 pp, high confidence |
| SWE-Rebench | 62.0% | estimated ± 6.5 pp, low confidence |
| PostTrainBench | 15.4% | estimated ± 3.7 pp, medium confidence |
| DAYJOB | 1.3% | estimated ± 3.2 pp, high confidence |
| DAYJOB Healthcare | 0.7% | estimated ± 4.2 pp, high confidence |
| BioMysteryBench (human-solvable) | 83.3% | estimated ± 0.6 pp, low confidence |
| HealthBench (raw) | 36.9% | estimated ± 6.0 pp, low confidence |
| HealthBench (length-adjusted) | 49.1% | estimated ± 3.4 pp, low confidence |
| HealthBench Professional | 38.2% | estimated ± 4.2 pp, low confidence |
| HealthBench Professional (raw) | 48.2% | estimated ± 4.5 pp, medium confidence |
| HLE-Verified | 15.3% | estimated ± 2.0 pp, low confidence |
| LABBench2 | 80.1% | estimated ± 1.8 pp, low confidence |
| ScarfBench | 25.2% | estimated ± 9.1 pp, low confidence |
| BlenderWorld (text-only) | 70.0% | estimated ± 2.8 pp, low confidence |
| Cookie-Bench | 76.9% | estimated ± 1.3 pp, low confidence |
| STEMBench | 2.5% | estimated ± 0.2 pp, medium confidence |
| SimpleQA | 42.1% | estimated ± 4.3 pp, low confidence |
| BabyVision w/ Python | 88.8% | estimated ± 3.8 pp, low confidence |
| OpenHarmony Bench v1.0 | 58.8% | estimated ± 2.4 pp, high confidence |
| SEC-Bench Pro | 33.7% | estimated ± 9.3 pp, medium confidence |
| BabyVision | 57.5% | estimated ± 9.1 pp, low confidence |
| MMVU | 76.3% | estimated ± 2.7 pp, medium confidence |
| PerceptionBench | 58.9% | estimated ± 4.9 pp, low confidence |
| MRCRv2 | 85.2% | estimated ± 11.3 pp, low confidence |
| MMLongBench-Doc | 61.3% | estimated ± 2.6 pp, low confidence |
| OCRBench V2 | 70.8% | estimated ± 3.3 pp, high confidence |
| BioBasicsBench | 33.9% | estimated ± 2.8 pp, low confidence |
| MMMU-Pro w/ Python | 78.0% | measured |
| ExploitBench Internal Port | 0.0% | estimated ± 1.4 pp, low confidence |