- Agents' Last Exam (Sol, xhigh, code mode) 53.6
- Artificial Analysis Intelligence Index v4.1 (Sol, max) 58.9
- Artificial Analysis Coding Agent Index v1.1 (Sol, max) 80
- Terminal-Bench 2.1 (Sol Ultra, 4 agents) 91.9
- DeepSWE v1.1 (Sol, max) 72.7
- BrowseComp (Sol Ultra, 4 agents) 92.2
- GDPval-AA v2 (Sol, max) 1748
- OSWorld 2.0 (Sol, max) 62.6
- AutomationBench (Sol, max) 18.1
- GeneBench Pro (Sol, max) 28.7
- LifeSciBench (Sol, max) 59.9
GPT-5.6 Sol
GPT-5.6 Sol ranks first on 10 of the 247 benchmarks it has scores on.
264
Benchmarks with a score
10
Ranked first, out of five or more models
54
In the top three
OpenAI Jul 2026
GPT-5.6 Sol
Closed weights Flagship
162.6
#12 of 271
Epoch Capabilities Index
- DeepSWE 72.7% #4 /28
- Agents' Last Exam 30.3% #9 /12
- Terminal-Bench 4.0 37.3% #11 /23
Where it ranks
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.
Knowledge and language
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SEA-HELM | 77.0 | #3 of 62 | Official leaderboard | |
| LMArena (Elo) | 1,485 | #19 of 333 | Official leaderboard | |
| Creative Writing v3 | 80.8 | #8 of 120 | Official leaderboard | |
| FACTS Parametric | 80.7% | #3 of 45 | Official leaderboard | |
| Wer wird Millionär? LLM Benchmark | 358333.0 | #6 of 76 | Official leaderboard | |
| Longform Creative Writing (EQ-Bench) | 81.7 | #11 of 119 | Official leaderboard | |
| FACTS Search V2 | 88.5% | #2 of 21 | Official leaderboard | |
| Extended NYT Connections | 93.8% | #10 of 97 | Official leaderboard | |
| FACTS Parametric V2 | 81.1% | #3 of 20 | Official leaderboard | |
| SimpleQA Verified | 69.2% | #9 of 56 | Official leaderboard | |
| Document Arena | 1,489 | #6 of 37 | Official leaderboard | |
| Global MMLU Lite | 94.1% | #11 of 43 | Official leaderboard | |
| FACTS Benchmark Suite | 56.5% | #12 of 40 | Official leaderboard | |
| Continual Learning Exam | 33.2% | #4 of 10 | Official leaderboard | |
| BullshitBench V2 | 48% | #49 of 122 | Official leaderboard | |
| EQ-Bench 4 | 1,236 | #13 of 32 | Official leaderboard | |
| HLE-Diamond | 41% | #8 of 18 | Official leaderboard | |
| BullshitBench | 27.3% | #55 of 107 | Official leaderboard | |
| Vectara Hallucination Leaderboard (2025 dataset) | 12.4% | #85 of 104 | Official leaderboard |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| CritPt | 32.3% | #2 of 160 | Epoch AI | |
| OTIS Mock AIME 2024-2025 | 100% | #4 of 202 | Epoch AI | |
| ARC-AGI-2 | 92.5% | #4 of 95 | Official leaderboard | |
| ProofBench | 77% | #3 of 59 | Official leaderboard | |
| GPQA Diamond | 93.5% | #16 of 219 | Epoch AI | |
| FrontierMath Tiers 1-3 v2 | 89.1% | #7 of 89 | Epoch AI | |
| ARC-AGI | 97.5% | #8 of 92 | Official leaderboard | |
| Verifying the Verifiers (Project APE CRED benchmark) | 99% | #2 of 23 | Official leaderboard | |
| DTBench | 92.5% | #20 of 178 | Official leaderboard | |
| FrontierMath Tier 4 v2 | 82.9% | #8 of 71 | Epoch AI | |
| Surface Evolver Bench | 93.1% | #3 of 26 | Epoch AI | |
| ResearchClawBench | 34.8 | #3 of 23 | Official leaderboard | |
| Leipzig Benchmark | 90% | #3 of 19 | Official leaderboard | |
| ExplorationBench | 88.1% | #2 of 10 | Official leaderboard | |
| SimpleBench | 64.8% | #26 of 100 | Official leaderboard | |
| ProofBench v1.1 | 83% | #14 of 51 | Official leaderboard | |
| Terminal-Bench-Science | 22.4% | #5 of 17 | Official leaderboard | |
| BixBench3 | 48.0 | #6 of 20 | Official leaderboard | |
| CritPt v2 | 74.9% | #4 of 11 | Official leaderboard | |
| Human Behavior Bench | 79.5 | #4 of 11 | Official leaderboard | |
| FrontierPhysics | 8.30% | #7 of 19 | Official leaderboard | |
| PaperBenchX | 9.68% | #4 of 10 | Official leaderboard | |
| MysteryMechanism | 33.3% | #12 of 29 | Official leaderboard | |
| LatchBio Capabilities v1.0 | 43.3% | #13 of 27 | Official leaderboard | |
| BTF-3 | 0.14 | #6 of 10 | Epoch AI | |
| FrontierMath Erdős | 0% | #9 of 14 | Epoch AI | |
| Prophet Arena | 84.7 | #15 of 18 | Official leaderboard | |
| ForecastBench | 65.8 | #81 of 86 | Official leaderboard | |
| SciUniverse | 9.40% | #6 of 6 | Official leaderboard |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| IOI (Vals) | 86.7% | #2 of 55 | Official leaderboard | |
| ALE-Bench | 2,177 | #5 of 121 | Epoch AI | |
| WeirdML | 89.4% | #6 of 112 | Official leaderboard | |
| FrontierCS | 76.4 | #1 of 14 | Official leaderboard | |
| AndroidBench | 90.8% | #3 of 32 | Official leaderboard | |
| IOI v2 (Vals) | 91.2% | #5 of 45 | Official leaderboard | |
| GSO-Bench | 76.5% | #4 of 30 | Official leaderboard | |
| DeepSWE | 72.7% | #4 of 28 | Official leaderboard | |
| WebDev Arena | 1,617 | #16 of 105 | Epoch AI | |
| Code Migration | 52.9% | #12 of 78 | Official leaderboard | |
| Senior SWE-Bench | 34.7% | #4 of 23 | Official leaderboard | |
| PostTrainBench | 36.2% | #2 of 11 | Epoch AI | |
| SWE-sweep | 4.70% | #2 of 10 | Official leaderboard | |
| Code Arena | 1,616 | #26 of 124 | Official leaderboard | |
| Image-to-WebDev Arena | 1,609 | #12 of 55 | Official leaderboard | |
| SWE-bench-Live | 70.3% | #3 of 13 | Official leaderboard | |
| Vibe Code Bench | 80.5% | #28 of 115 | Official leaderboard | |
| PostTrainBench 1.1 | 36.2% | #3 of 11 | Official leaderboard | |
| FrontierCode | 47.5% | #13 of 46 | Epoch AI | |
| SWE Atlas | 46% | #7 of 22 | Official leaderboard | |
| SWE Atlas Test Writing | 45.9% | #7 of 22 | Official leaderboard | |
| ProgramBench | 1.00% | #8 of 25 | Official leaderboard | |
| GBAEval | 52.6% | #7 of 21 | Official leaderboard | |
| RSI-Exam | 43.3% | #6 of 18 | Official leaderboard | |
| Artificial Analysis Coding Agent Index v1.1 | 80.0 | #1 of 3 | Model card | |
| TasteVal | 59.0× | #7 of 20 | Official leaderboard | |
| SWE Refactor Bench | 12.4% | #3 of 8 | Official leaderboard | |
| FrontierCode 2.0 | 27.6% | #10 of 26 | Official leaderboard | |
| SWE-rebench | 62.3% | #5 of 13 | Official leaderboard | |
| MLS-Bench | 46.2% | #6 of 15 | Official leaderboard | |
| FrontierSWE | 32.2% | #9 of 21 | Epoch AI | |
| AutoResearchExam | 0.51 | #4 of 9 | Official leaderboard | |
| Vibe Code Bench 1-100 | 20% | #12 of 26 | Official leaderboard | |
| Vals RSI Index | 0.24 | #12 of 25 | Official leaderboard | |
| WeirdML v3 | 15% | #12 of 25 | Official leaderboard | |
| GameDevBench | 63.7% | #5 of 10 | Official leaderboard | |
| ReviewBench | 70.2% | #2 of 4 | Official leaderboard | |
| CursorBench | 41.7% | #11 of 19 | Epoch AI | |
| PostTrainBench 1.2 | 31% | #10 of 17 | Official leaderboard | |
| SWE-Bench Pro V2 | 95.5% | #10 of 16 | Official leaderboard | |
| SWE-Bench Pro V2 Hard | 82.4% | #10 of 16 | Official leaderboard | |
| Brokk Power Ranking | 7.10 | #13 of 19 | Official leaderboard | |
| Vision2Web | 73.5% | #7 of 10 | Official leaderboard | |
| OpenDesign Arena | 77.6 | #14 of 19 | Official leaderboard | |
| MirrorCode | 20% | #24 of 28 | Epoch AI |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Chess Puzzles | 68% | #2 of 83 | Epoch AI | |
| BrowseComp | 92.2% | #2 of 77 | Model card | |
| LMArena Search Arena | 1,258 | #1 of 35 | Official leaderboard | |
| Game Arena Poker | 30.3 | #1 of 31 | Official leaderboard | |
| Terminal-Bench 2.1 | 91.9% | #2 of 53 | Model card | |
| LLM Chess | 1,550 | #5 of 124 | Official leaderboard | |
| Long-Horizon-Terminal-Bench (LHTB) | 60% | #2 of 25 | Official leaderboard | |
| BALROG | 60% | #3 of 35 | Official leaderboard | |
| Mystery Game Puzzles | 58% | #8 of 80 | Epoch AI | |
| MobileWorld | 70.1% | #2 of 18 | Official leaderboard | |
| Long-Horizon-Terminal-Bench 2.0 | 67.2% | #2 of 15 | Official leaderboard | |
| Game Arena Word Association | 361 | #5 of 36 | Official leaderboard | |
| CEO-Bench | $11,313,982 | #3 of 19 | Official leaderboard | |
| τ³-bench Banking | 46.9% | #4 of 25 | Official leaderboard | |
| Terminal-Bench 3.0 | 34.6% | #2 of 11 | Official leaderboard | |
| LisanBench | 4699.0 | #25 of 133 | Official leaderboard | |
| PinchBench 2.0 | 84.2% | #12 of 57 | Official leaderboard | |
| Game Arena Dark Hex | 341 | #8 of 38 | Official leaderboard | |
| Game Arena Ultimate Tic-Tac-Toe | 449 | #8 of 36 | Official leaderboard | |
| OSWorld 2.0 | 27.3% | #2 of 9 | Epoch AI | |
| Game Arena Word Art | 1,074 | #8 of 35 | Official leaderboard | |
| Game Arena Bargaining | 224 | #9 of 36 | Official leaderboard | |
| Vals Web Search Index | 49.1% | #1 of 4 | Official leaderboard | |
| Agent Arena | 0.05 | #14 of 53 | Official leaderboard | |
| Game Arena Lines of Action | 638 | #10 of 36 | Official leaderboard | |
| ApprenticeBench | 26% | #5 of 18 | Official leaderboard | |
| Game Arena Chess | 1,134 | #11 of 39 | Official leaderboard | |
| Kaggle Game Arena | 352 | #11 of 39 | Official leaderboard | |
| RecreationBench | 42.1% | #3 of 10 | Official leaderboard | |
| EBR-bench | 44.8% | #9 of 28 | Epoch AI | |
| MCP Atlas | 81.8% | #11 of 34 | Official leaderboard | |
| Brood War Bench | 72.2% | #2 of 6 | Official leaderboard | |
| Game Arena Coin Game | 415 | #12 of 35 | Official leaderboard | |
| Game Arena Clobber | 340 | #13 of 36 | Official leaderboard | |
| ARC-AGI-3 | 7.78% | #9 of 24 | Official leaderboard | |
| Time Horizon Index: KSP | 23.8% | #5 of 13 | Official leaderboard | |
| Game Arena Go | 1,011 | #14 of 36 | Official leaderboard | |
| Game Arena Dots and Boxes | 933 | #16 of 37 | Official leaderboard | |
| Game Arena Nine Men's Morris | 637 | #16 of 36 | Official leaderboard | |
| Game Arena Reversi | 590 | #16 of 36 | Official leaderboard | |
| MyPCBench | 55.4% | #4 of 9 | Official leaderboard | |
| SnakeBench | 25 | #86 of 183 | Official leaderboard | |
| Terminal-Bench 4.0 | 37.3% | #11 of 23 | Official leaderboard | |
| Game Arena Four in a Row | 358 | #19 of 38 | Official leaderboard | |
| SaaS-Bench | 29.2% | #2 of 4 | Official leaderboard | |
| CUA-bench | 8.33% | #5 of 9 | Official leaderboard | |
| Game Arena Checkers | 881 | #23 of 38 | Official leaderboard | |
| AutomationBench | 18.1% | #36 of 41 | Model card | |
| OSWorld 2.0 (Aug 2026) | 27.3% | #3 of 3 | Official leaderboard |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| E-Commerce Bench | 14.3 | #1 of 19 | Official leaderboard | |
| ProfBench | 60.4% | #4 of 72 | Official leaderboard | |
| LegalBench | 87% | #9 of 138 | Official leaderboard | |
| PhysicianBench | 50% | #1 of 13 | Official leaderboard | |
| PolicyBench | 93.1% | #4 of 50 | Official leaderboard | |
| Excel Modeling Benchmark | 72.3% | #7 of 76 | Official leaderboard | |
| Legal Research Bench | 48.1% | #11 of 88 | Official leaderboard | |
| Vending-Bench 2 | $9,619 | #10 of 66 | Official leaderboard | |
| TaxCalcBench v2 | 68% | #5 of 26 | Official leaderboard | |
| RCM-Bench-1 | 91.2% | #4 of 20 | Official leaderboard | |
| OR-License | 84.5% | #2 of 10 | Official leaderboard | |
| PRBench | 50.5% | #9 of 44 | Official leaderboard | |
| BlenderWorld | 67.7 | #3 of 13 | Official leaderboard | |
| TaxEval v2 | 74.8% | #30 of 129 | Official leaderboard | |
| GDPval-AA v2 | 1,748 | #7 of 26 | Model card | |
| Vending-Bench Arena | $7,821 | #10 of 32 | Official leaderboard | |
| MortgageTax | 67.3% | #27 of 86 | Official leaderboard | |
| MedScribe | 85.2% | #31 of 98 | Official leaderboard | |
| Tax Agent Bench | 31.1% | #25 of 79 | Official leaderboard | |
| CorpFin v2 | 64.4% | #39 of 119 | Official leaderboard | |
| Public Benefits Bench | 66.5% | #18 of 50 | Official leaderboard | |
| CommerceAgentBench | 51.4% | #5 of 13 | Official leaderboard | |
| Vals Finance Agent v2 | 53.8% | #31 of 80 | Official leaderboard | |
| MedCode | 44% | #41 of 96 | Official leaderboard | |
| FrontierFinance | 46.8% | #7 of 16 | Official leaderboard | |
| JobBench | 45.4% | #7 of 16 | Official leaderboard | |
| Short Form Bench | $2,410 | #4 of 9 | Official leaderboard | |
| HealthBench Consensus | 95.5% | #8 of 17 | Model card | |
| APEX-Agents | 51.4% | #29 of 56 | Epoch AI | |
| HealthBench Hard | 33.1% | #11 of 21 | Model card | |
| HealthBench Professional | 60.5% | #14 of 25 | Model card | |
| Harvey LAB (Legal Agent Benchmark) | 2.50% | #47 of 80 | Official leaderboard | |
| Parametric CAD Bench | 30% | #7 of 11 | Official leaderboard | |
| HealthBench | 57% | #21 of 30 | Model card | |
| Agents' Last Exam | 30.3% | #9 of 12 | Official leaderboard | |
| FoodTruck Bench | $53,229 | #8 of 10 | Official leaderboard |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| ZeroBench | 30% | #4 of 66 | Official leaderboard | |
| ParseBench | 74.5% | #4 of 42 | Official leaderboard | |
| Vision Arena | 1,286 | #15 of 136 | Official leaderboard | |
| BabyVision | 88.9% | #3 of 27 | Official leaderboard | |
| VoxelBench | 2,288 | #7 of 56 | Official leaderboard | |
| GDP.pdf | 30.7% | #5 of 39 | Epoch AI | |
| ClockBench | 68.3% | #5 of 37 | Official leaderboard | |
| Code2Video Benchmark | 1,548 | #4 of 19 | Official leaderboard | |
| Furniture Assembly Benchmark | 56.7% | #10 of 35 | Epoch AI | |
| ArtifactArena | 1,165 | #7 of 21 | Official leaderboard | |
| Drone-Bench | 76.1% | #7 of 19 | Official leaderboard | |
| Blueprint-Bench 2 | 33.6% | #13 of 35 | Official leaderboard | |
| EnigmaEval | 37.1% | #2 of 5 | Official leaderboard | |
| GeoBench | 73% | #25 of 48 | Official leaderboard | |
| Humanity's Sixth Sense | 30% | #18 of 29 | Official leaderboard | |
| VibeVideoBench | 3.32 | #6 of 9 | Official leaderboard | |
| OmniParseBench | 87.8 | #5 of 6 | Official leaderboard | |
| 1H-VideoQA | 62.4% | #17 of 20 | Official leaderboard | |
| RoboDojo-RC Tier 1 | 14.9% | #8 of 9 | Official leaderboard |
Safety and security
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| CyberBench (Vals) | 88.1% | #1 of 26 | Official leaderboard | |
| Human Pathogen Capabilities Test v2 (HPCT v2) | 70.6% | #5 of 71 | Official leaderboard | |
| KORA | 68.7% | #3 of 42 | Official leaderboard | |
| CVE-Bench | 94.1% | #1 of 14 | Model card | |
| ProtocolQA Open-Ended | 44.1% | #2 of 25 | Model card | |
| CyberBench v1.1 (Vals) | 76.3% | #4 of 43 | Official leaderboard | |
| scBench-Long | 29.6% | #3 of 27 | Official leaderboard | |
| Human Pathogen Capabilities Test | 68.4% | #1 of 8 | Model card | |
| Agent JSK | 100% | #1 of 7 | Model card | |
| Molecular Biology Capabilities Test | 60% | #1 of 7 | Model card | |
| SRE-Bench | 30.5% | #5 of 33 | Official leaderboard | |
| TroubleshootingBench | 48% | #3 of 19 | Model card | |
| Tacit Knowledge and Troubleshooting | 82.2% | #5 of 29 | Model card | |
| SpatialBench-Long | 37.5% | #5 of 27 | Official leaderboard | |
| Cyber Range | 83.3% | #2 of 10 | Model card | |
| DeepsecBench | 35.3 | #6 of 28 | Official leaderboard | |
| ExploitGym | 33.7% | #4 of 18 | Official leaderboard | |
| Multimodal Troubleshooting Virology | 55.5% | #6 of 26 | Model card | |
| TxBench-Preclinical-Pharmacology | 61.3% | #7 of 28 | Official leaderboard | |
| Hard-negative protein binding prediction | 7.60% | #1 of 4 | Model card | |
| DrugDiscoveryBench | 56.5% | #5 of 19 | Official leaderboard | |
| CyberGym | 83.6% | #9 of 33 | Model card | |
| Dictatorship Eval | 85.4% | #6 of 20 | Official leaderboard | |
| EpiBench | 25.8% | #9 of 27 | Official leaderboard | |
| WellnessBench | 57% | #4 of 12 | Official leaderboard | |
| Last Secure Code Benchmark | 24.7% | #2 of 6 | Official leaderboard | |
| Capture the Flag (Internal) | 97.1% | #1 of 3 | Model card | |
| World-Class Bio | 68.3% | #1 of 3 | Model card | |
| DistressBench | 73.6% | #15 of 42 | Official leaderboard | |
| FORTRESS | 20.1% | #26 of 66 | Official leaderboard | |
| OncologyBench | 18.7% | #12 of 30 | Official leaderboard | |
| First-Person Fairness evaluation | 0.98% | #4 of 10 | Model card | |
| PCT (Polarized Contrastive Pairs) | 41.7% | #19 of 46 | Official leaderboard | |
| CyScenarioBench | 28% | #6 of 14 | Model card | |
| BioSecBench-Surveillance | 47.3% | #12 of 27 | Official leaderboard | |
| TxBench-Oligo-Discovery | 37.2% | #13 of 28 | Official leaderboard | |
| Virology Capabilities Test (VCT) | 53.5% | #13 of 28 | Model card | |
| VariantBench | 35.3% | #13 of 27 | Official leaderboard | |
| Destructive action avoidance | 83% | #4 of 8 | Model card | |
| AAV Capsid Packaging Prediction (OpenAI) | 0.53 | #3 of 6 | Model card | |
| GeneBench Pro | 28.7% | #2 of 4 | Model card | |
| LifeSciBench | 59.9% | #2 of 4 | Model card | |
| UK AISI cyber ranges | 70% | #2 of 4 | Model card | |
| ReproBAIT | 85% | #1 of 2 | Model card | |
| UK AISI narrow cyber tasks | 95% | #1 of 2 | Model card | |
| TxBench-Antibody-Discovery | 37.3% | #15 of 28 | Official leaderboard | |
| BioSecBench-Function | 35.5% | #15 of 27 | Official leaderboard | |
| UK AISI unprompted sabotage evaluation | 0% | #3 of 5 | Model card | |
| MetagenomicsBench | 31.3% | #17 of 26 | Official leaderboard | |
| Prompt injection attacks in search and function calling | 91% | #2 of 3 | Model card | |
| Instruction hierarchy attacks | 99.9% | #5 of 7 | Model card | |
| CoT-Control | 1.30% | #6 of 8 | Model card | |
| DNA sequence design for transcription factor binding | 13.7% | #3 of 4 | Model card | |
| Prompt injection robustness | 96.2% | #7 of 7 | Model card | |
| UK AISI sabotage continuation evaluation | 1.00% | #4 of 4 | Model card | |
| FrontierCyber | 9.64% | #2 of 2 | Model card |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LMCA | 59.2% | #9 of 139 | Epoch AI | |
| Conceptual Reasoning Index | 68.7 | #11 of 149 | Official leaderboard | |
| SAGE (Vals) | 52.6% | #6 of 81 | Official leaderboard | |
| Political Science LLM Benchmark | 69.2% | #3 of 29 | Official leaderboard | |
| Vals Multimodal Index | 72.6% | #4 of 33 | Official leaderboard | |
| CAIS Text Capabilities Index | 48.8 | #10 of 66 | Official leaderboard | |
| ACCoRD | 75.9 | #23 of 150 | Official leaderboard | |
| CAIS Vision Capabilities Index | 66.4 | #11 of 53 | Official leaderboard | |
| Artificial Analysis Intelligence Index v4.1 | 58.9 | #2 of 7 | Model card | |
| CAIS Risk Index | 44.5 | #21 of 65 | Official leaderboard | |
| Vals Science Index | 41.5% | #10 of 23 | Official leaderboard |
OpenAI's models over time
OpenAI's models in color, against every other model on the Epoch Capabilities Index, by release date.
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.
Its system cards
Every result in the longer documents the lab published with it. 51 of them aren't in the launch post.
GPT-5.6 Preview System Card 25 Jun 2026
Health
Hallucination
Safety
- OpenAI Production Benchmarks and safe-completion evals
- Disallowed Content Evaluations with Representative Prompts
- Image input safety evaluations
- Destructive action avoidance 83
- User Confirmations During Computer Use
- Biology Model Refusal Evaluation
- OpenAI cyber safety evaluations (incl. Daybreak Blue)
Robustness
Bias
Alignment
Monitorability
Bio and chem risk
- Multimodal Troubleshooting Virology 55.5
- ProtocolQA Open-Ended 43.5
- Tacit Knowledge and Troubleshooting
- TroubleshootingBench 48
- AAV Capsid Packaging Prediction (OpenAI) 0.529
- Hard-negative protein binding prediction 7.6
- DNA sequence design for transcription factor binding 13.7
- Virology Capabilities Test (VCT) 53.5
- Molecular Biology Capabilities Test 60
- Human Pathogen Capabilities Test 68.4
- World-Class Bio 68.3
- ReproBAIT 85
Cyber risk
GPT-5.6 System Card 9 Jul 2026
Health
Hallucination
Safety
- OpenAI Production Benchmarks and safe-completion evals
- Disallowed Content Evaluations with Representative Prompts
- Image input safety evaluations
- Destructive action avoidance 83
- User Confirmations During Computer Use
- Biology Model Refusal Evaluation
- OpenAI cyber safety evaluations (incl. Daybreak Blue)
Robustness
Bias
Alignment
Monitorability
Bio and chem risk
- Multimodal Troubleshooting Virology 55.5
- ProtocolQA Open-Ended 43.5
- Tacit Knowledge and Troubleshooting
- TroubleshootingBench 48
- AAV Capsid Packaging Prediction (OpenAI) 0.529
- Hard-negative protein binding prediction 7.6
- DNA sequence design for transcription factor binding 13.7
- Virology Capabilities Test (VCT) 53.5
- Molecular Biology Capabilities Test 60
- Human Pathogen Capabilities Test 68.4
- World-Class Bio 68.3
- ReproBAIT 85
Cyber risk
GPT-5.6 – August Updates 6 Aug 2026
Health
Hallucination
Safety
Robustness
Bio and chem risk
Cyber risk
- Capture the Flag (Internal) 97.1
- CVE-Bench 94.1
- Cyber Range 83.3