OpenAIGPT
GPT-5.1
OpenAI · GPT. Released Nov 12, 2025.
GPT-5.1 is a model from OpenAI in the GPT family, released Nov 12, 2025. evals.report tracks 37 reported GPT-5.1 benchmark scores across FrontierMath, ARC-AGI-1, LiveCodeBench Pro, GPQA Diamond, SWE-bench Verified, Humanity's Last Exam, MMMU-Pro, AIME (OTIS Mock), and 29 more — each shown with its benchmark, metric, source status, and date, and never combined into a single ranking.
37 results
Benchmark results 37
Compare this model| Benchmark | Category | Score | Metric | Status | Date | |
|---|---|---|---|---|---|---|
| FrontierMath | Reasoning | 31.03% | accuracy | Official | Nov 12, 2025 | Details |
| ARC-AGI-1 | Reasoning | 72.83% | accuracy | Official | Nov 12, 2025 | Details |
| LiveCodeBench Pro | Coding | 2269 | Codeforces Elo | Official | Nov 12, 2025 | Details |
| GPQA Diamond | Reasoning | 87.6% | accuracy | Official | Nov 12, 2025 | Details |
| SWE-bench Verified | Coding | 68.0% | % resolved | Official | Nov 12, 2025 | Details |
| Humanity's Last Exam | Reasoning | 27.2% | accuracy | Official | Nov 12, 2025 | Details |
| MMMU-Pro | Multimodal | 79.0% | accuracy | Official | Nov 12, 2025 | Details |
| AIME (OTIS Mock) | Reasoning | 88.6% | accuracy | Official | Nov 12, 2025 | Details |
| SimpleQA Verified | Other | 48.9% | accuracy | Official | Nov 12, 2025 | Details |
| ARC-AGI-2 | Reasoning | 17.64% | accuracy | Official | Nov 12, 2025 | Details |
| LMArena | Chat preference | 1422 | source-defined rating | Official | Nov 12, 2025 | Details |
| WeirdML | Coding | 60.8% | average accuracy | Official | Nov 12, 2025 | Details |
| Artificial Analysis Intelligence Index | Reasoning | 47.7 | Index | Unverified | Nov 12, 2025 | Details |
| Epoch Capabilities Index | Reasoning | 149.7 | Index | Official | Nov 12, 2025 | Details |
| MMLU-Pro | Reasoning | 87.0% | accuracy | Verified | Nov 12, 2025 | Details |
| τ²-bench (Telecom) | Tool use | 95.6% | pass^1 | Unverified | Nov 12, 2025 | Details |
| GDPval | Agents | 1227 | Elo | Official | Nov 12, 2025 | Details |
| LiveCodeBench | Coding | 86.8% | Pass@1 | Unverified | Nov 12, 2025 | Details |
| SciCode | Coding | 43.3% | accuracy | Unverified | Nov 12, 2025 | Details |
| MMMU (Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark) | Multimodal | 85.4% | accuracy | Unverified | Nov 12, 2025 | Details |
| MultiChallenge | Reasoning | 63.41% | accuracy | Verified | Nov 12, 2025 | Details |
| OpenAI-MRCR v2 (Multi-Round Coreference Resolution) | Reasoning | 61.6% | accuracy (mean SequenceMatcher similarity) | Unverified | Nov 12, 2025 | Details |
| Global-MMLU | Reasoning | 90.6% | accuracy | Unverified | Nov 12, 2025 | Details |
| WebDev Arena | Chat preference | 1391 | Elo | Verified | Nov 12, 2025 | Details |
| Search Arena | Chat preference | 1199 | Elo | Verified | Nov 12, 2025 | Details |
| Design Arena | Chat preference | 1216 | Elo | Verified | Nov 12, 2025 | Details |
| MASK (Model Alignment between Statements and Knowledge) | Other | 86.33 | Honesty score | Verified | Nov 12, 2025 | Details |
| ScreenSpot-Pro | Multimodal | 3.5% | accuracy | Verified | Nov 12, 2025 | Details |
| FACTS Grounding | Reasoning | 50.0% | Grounding accuracy | Verified | Nov 12, 2025 | Details |
| EnigmaEval | Reasoning | 11.23% | accuracy | Verified | Nov 12, 2025 | Details |
| ZeroBench | Multimodal | 5.0% (pass@5) | accuracy | Verified | Nov 12, 2025 | Details |
| FrontierMath Tier 4 | Reasoning | 12.5% | accuracy | Official | Nov 12, 2025 | Details |
| Vectara Hallucination Leaderboard | Other | 12.1% | Hallucination Rate | Official | Nov 12, 2025 | Details |
| Gray Swan Arena (Agent Red-Teaming / Indirect Prompt Injection) | Agents | 2.5% | Attack Success Rate (ASR) | Verified | Nov 12, 2025 | Details |
| Vibe Code Bench | Coding | 24.61% | Overall accuracy | Verified | Nov 12, 2025 | Details |
| GSO: Software Optimization Benchmark for SWE-Agents | Coding | 13.73% | Opt@1 | Official | Nov 12, 2025 | Details |
| MultiNRC | Reasoning | 49.00% | accuracy | Official | Nov 12, 2025 | Details |