Overall score
66.4
Rank #9 · A confidence
OpenAI · stable
OpenAI's documented GPT-5.6 model positioned between frontier capability and API cost.
Overall score
66.4
Rank #9 · A confidence
Capability
61.7
27 eligible benchmark families
Coverage
100%
Confidence A
Benchmarks scored
32
37 raw result rows
Model specification
Identity, configuration and price are preserved as dated source records.
Category profile
A category remains Not evaluated until an eligible source-qualified result is attached.
Benchmark scores
Best published score per benchmark. Open any row for the full per-benchmark leaderboard.
| Benchmark | Category | Score | Date |
|---|---|---|---|
| BenchLM Math priorRanking weighted | mathematics | 95.3% | 2026-07-15 |
| GPQA DiamondRanking weighted | reasoning | 92.9% | 2026-07-15 |
| Terminal-BenchRanking weighted | agents | 88.01% | 2026-07-15 |
| BrowseCompRanking weighted | agents | 87.5% | 2026-07-15 |
| τ-benchRanking weighted | agents | 86.3% | 2026-07-15 |
| τ²-Bench TelecomRanking weighted | agents | 86.2573% | 2026-07-15 |
| FrontierMathRanking weighted | mathematics | 84.9% | 2026-07-15 |
| MMMU-ProRanking weighted | multimodal | 80.7% | 2026-07-15 |
| BenchLM Knowledge priorRanking weighted | research | 75.6% | 2026-07-15 |
| AA Long Context ReasoningRanking weighted | research | 74% | 2026-07-15 |
| BenchLM Multimodal priorRanking weighted | multimodal | 72.5% | 2026-07-15 |
| IFBench | instruction following | 71.2% | 2026-07-15 |
| IFEval | instruction following | 71.2% | 2026-07-15 |
| BenchLM Agentic priorRanking weighted | agents | 67.83% | 2026-07-15 |
| BenchLM Coding priorRanking weighted | coding | 66.59% | 2026-07-15 |
| CursorBenchRanking weighted | coding | 64.9% | 2026-07-15 |
| SWE-Bench ProRanking weighted | coding | 63.4% | 2026-07-15 |
| Long-Horizon-Terminal-BenchRanking weighted | agents | 57.6% | 2026-07-15 |
| Terminal-Bench HardRanking weighted | agents | 57.5758% | 2026-07-15 |
| GDPval-AA v2Ranking weighted | agents | 54.6475% | 2026-07-15 |
| SciCodeRanking weighted | coding | 53.9% | 2026-07-15 |
| ToolathlonRanking weighted | agents | 53.1% | 2026-07-15 |
| ITBench-AARanking weighted | agents | 51.0358% | 2026-07-15 |
| OSWorldRanking weighted | agents | 50.2% | 2026-07-15 |
| AA AutomationBenchRanking weighted | agents | 45.6% | 2026-07-15 |
| Humanity's Last ExamRanking weighted | reasoning | 41.8% | 2026-07-15 |
| τ³-BankingRanking weighted | agents | 31.7526% | 2026-07-15 |
| CritPtRanking weighted | reasoning | 30% | 2026-07-15 |
| ExploitGym | agents | 23.2% | 2026-07-15 |
| Harvey LAB-AARanking weighted | agents | 2.5% | 2026-07-15 |
| ARC-AGI-3 | reasoning | 0.8% | 2026-07-15 |
| AA-OmniscienceRanking weighted | research | -0.2167 index | 2026-07-15 |
Showing the best of 37 raw result rows across 32 benchmark families.
Price and efficiency
Operational values are shown only when the source publishes a comparable measurement or a valid derivation.
Verification summary
News and X posts are structurally excluded from this model's evidence graph.
Capability ranking is complete for the current snapshot. The last source check for this exact model was 2026-07-14.
Related models
Exact variants remain separate even when they share a provider or family.