Claude Opus 5 vs GPT-6.2 Sol
GPT-6.2 Sol leads by 3.0 capability points (≈95% interval +0.5 to +5.6): a clear lead over Claude Opus 5 · P(higher) ≈ 99%
Wins 29 of 43 shared benchmarks (12 lost, 2 tied)
Capability vs. relative cost
Claude Opus 5GPT-6.2 SolClosed weightsOpen weights
Benchmark wins12 – 29 · 43 shared, 2 tied
Comparing matching setups only (11 benchmarks excluded) · suites counted once
Area← Claude Opus 5GPT-6.2 Sol →
Data: Artificial Analysis, Vals AI, Arena and 12 othersatlas.village.dev
Margins
Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is GPT-6.2 Sol minus Claude Opus 5.
| Terminal-Bench-Science 0.1 | Science › Multidisciplinary | 28.6 | 67.6 | +39.0 |
| EnterpriseBench: CoreCraft | Tool & computer use › Customer service & operations | 68.7 | 36.4 | −32.3 |
| Terminal-Bench Science 0.1 (Vals) | Science › Multidisciplinary | 27.1 | 58.6 | +31.4 |
| ExploitBench - release-reported exploit progress | Software › Cybersecurity | 70.0 | 100.0 | +30.0 |
| Blueprint-Bench 2 | Perception & media › Spatial & visual reasoning | 0.3 | 0.6 | +26.6 |
| Humanity's Sixth Sense | Perception & media › Spatial & visual reasoning | 30.5 | 56.8 | +26.3 |
| DeepsecBench-AA (Artificial Analysis) | Software › Cybersecurity | 25.9 | 51.6 | +25.7 |
| Medical Long Context Reasoning (MLCR-AA, October 2026 methodology) | Health & medicine | 13.9 | 37.2 | +23.3 |
| Humanity's Last Exam Diamond | Knowledge & reasoning › General knowledge | 44.8 | 64.0 | +19.2 |
| EnterpriseOps-Gym-AA | Tool & computer use › Customer service & operations | 47.5 | 28.9 | −18.6 |
| Legal Research Bench | Professional › Legal | 55.3 | 37.0 | −18.3 |
| DeepsecBench — Revalidated V1 (233 issues) | Software › Cybersecurity | 32.3 | 47.3 | +15.0 |
| AutomationBench-AA 1.0.6 | Tool & computer use › Customer service & operations | 56.6 | 71.3 | +14.7 |
| AutomationBench 1.0.6 (Zapier) | Tool & computer use › Customer service & operations | 26.9 | 41.3 | +14.3 |
| Tax Agent Bench | Professional › Finance, accounting & tax | 76.7 | 62.9 | −13.7 |
| FrontierCode 2.0 (Artificial Analysis) | Software › Software engineering | 31.2 | 44.4 | +13.2 |
| Artificial Analysis Intelligence Index v5.0.1 | Composite indexes | 44.2 | 57.3 | +13.1 |
| TaxCalcBench TY25 | Professional › Finance, accounting & tax | 20.0 | 32.0 | +12.0 |
| Terminal-Bench 4.0 (Artificial Analysis) | Software › Software engineering | 49.0 | 60.1 | +11.1 |
| SimpleBench | Knowledge & reasoning › General knowledge | 80.6 | 69.9 | −10.7 |
| SWE-sweep | Software › Software engineering | 1.3 | 11.6 | +10.3 |
| LMArena Agent Arena Net Improvement (Steering Burden) | Tool & computer use › General agents | 6.3 | 11.3 | +9.8 |
| Vals Science Index | Science › Multidisciplinary | 47.9 | 57.8 | +9.8 |
| DAYJOB: Finance (October 2026 rescoring) | Professional › Finance, accounting & tax | 23.8 | 32.8 | +9.0 |
| GDP.pdf (Artificial Analysis, All-pass) | Professional › General knowledge work | 21.6 | 30.0 | +8.4 |
| CritPt v2 (agentic, avg@5) | Science › Physical sciences | 71.4 | 78.6 | +7.1 |
| LMArena Agent Arena Praise vs Complaint | Tool & computer use › General agents | 10.3 | 23.3 | +6.5 |
| LMArena Agent Arena Steering Burden | Tool & computer use › General agents | 2.4 | 13.4 | +5.5 |
| DAYJOB: Healthcare (October 2026 rescoring) | Health & medicine | 22.0 | 17.2 | −4.8 |
| ARC-AGI-2 | Knowledge & reasoning › Abstract reasoning | 90.4 | 85.8 | −4.6 |
| Chartography (November 2026 rescoring) | Perception & media › Documents & charts | 84.4 | 87.7 | +3.3 |
| TxBench Preclinical Pharmacology V1 Clean Reruns | Science › Life sciences | 63.0 | 65.3 | +2.3 |
| AA-Briefcase v1.1 Elo | Professional › General knowledge work | 1654 | 1612 | −2.3 |
| LMArena Agent Arena Bash Reliability | Tool & computer use › General agents | 8.7 | 4.3 | −2.2 |
| BioMysteryBench (Vals Terminus 2) | Science › Life sciences | 79.3 | 77.8 | −1.5 |
| AA-Omniscience v1.1 Index | Knowledge & reasoning › General knowledge | 37.6 | 40.4 | +1.4 |
| ARC-AGI-3 (Semi-Private) | Games | 30.2 | 31.1 | +0.9 |
| Vibe Code Bench v1.1 | Software › Frontend & design | 88.4 | 87.6 | −0.8 |
| MysteryMechanism | Math | 37.4 | 36.9 | −0.5 |
| LMArena Agent Arena Confirmed Success | Tool & computer use › General agents | 10.1 | 10.9 | +0.4 |
| LMArena Agent Arena Tool Hallucination | Tool & computer use › Tool calling | 1.0 | 1.2 | +0.1 |
| AA-AnalystAgent (pass^5) | Professional › General knowledge work | 53.8 | 53.8 | 0.0 |
| ARC-AGI-1 Semi-Private | Knowledge & reasoning › Abstract reasoning | 97.5 | 97.5 | 0.0 |
Data: Artificial Analysis, Vals AI, Arena and 12 othersatlas.village.dev