Claude Sonnet 5.5 vs GPT-6.2 Sol
GPT-6.2 Sol leads by 0.4 capability points (≈95% interval −2.3 to +3.1): statistically tied · P(higher) ≈ 61%
Wins 37 of 69 shared benchmarks (32 lost)
Capability vs. relative cost
Claude Sonnet 5.5GPT-6.2 SolClosed weightsOpen weights
Benchmark wins32 – 37 · 69 shared
Comparing matching setups only (33 benchmarks excluded) · suites counted once
Area← Claude Sonnet 5.5GPT-6.2 Sol →
Data: Artificial Analysis, Surge AI, Arena and 16 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 Sonnet 5.5.
| AA-PersonalityBench Informativeness index | Language & writing › Style & personality | 84.8 | 37.7 | −47.5 |
| AA-PersonalityBench Transparency index | Language & writing › Style & personality | 100 | 56.9 | −43.5 |
| AA-PersonalityBench Organization index | Language & writing › Style & personality | 84.8 | 45.7 | −39.5 |
| EnterpriseBench: CoreCraft | Tool & computer use › Customer service & operations | 70.8 | 36.4 | −34.4 |
| AA-PersonalityBench Imagination index | Language & writing › Style & personality | 84.8 | 60.1 | −24.9 |
| ARC-AGI-3 (Semi-Private) | Games | 72.4 | 97.4 | +24.9 |
| AA-PersonalityBench Conviction index | Language & writing › Style & personality | 58.1 | 79.2 | +21.3 |
| AA-PersonalityBench Expressiveness index | Language & writing › Style & personality | 73.3 | 55.3 | −18.2 |
| Blueprint-Bench 2 | Perception & media › Spatial & visual reasoning | 0.4 | 0.6 | +17.7 |
| DAYJOB: Healthcare (October 2026 rescoring) | Health & medicine | 34.0 | 17.2 | −16.8 |
| Ramp SWE-Bench v2 (Bash harness) | Software › Software engineering | 57.6 | 72.4 | +14.8 |
| sudo L7 (October 2026 rescoring) | Software › Software engineering | 50.6 | 36.1 | −14.5 |
| Terminal-Bench-Science 0.1 | Science › Multidisciplinary | 53.3 | 67.6 | +14.3 |
| TaxCalcBench TY25 | Professional › Finance, accounting & tax | 18.0 | 32.0 | +14.0 |
| Chartography | Perception & media › Documents & charts | 61.6 | 75.1 | +13.5 |
| AA-PersonalityBench Warmth index | Language & writing › Style & personality | 73.3 | 60.1 | −13.4 |
| HANDBOOK.md Agents v1.1 | Tool & computer use › Customer service & operations | 47.7 | 34.5 | −13.2 |
| Agents' Last Exam v1.1 Score | Professional › General knowledge work | 60.4 | 73.0 | +12.6 |
| AA-PersonalityBench Slop index | Language & writing › Style & personality | 27.7 | 15.4 | −12.4 |
| MysteryMechanism | Math | 49.1 | 36.9 | −12.2 |
| Terminal-Bench Science 0.1 (Vals) | Science › Multidisciplinary | 47.1 | 58.6 | +11.4 |
| AA-Briefcase v1.1 Elo | Professional › General knowledge work | 1815 | 1612 | −11.1 |
| Humanity's Last Exam Diamond | Knowledge & reasoning › General knowledge | 53.3 | 64.0 | +10.7 |
| DeepsecBench — Revalidated V1 (233 issues) | Software › Cybersecurity | 37.5 | 47.3 | +9.8 |
| LMArena Agent Arena Net Improvement (Steering Burden) | Tool & computer use › General agents | 15.6 | 11.3 | −8.3 |
| SWE Atlas v1.1 - Overall | Software › Software engineering | 68.3 | 60.5 | −7.8 |
| DAYJOB: Finance (October 2026 rescoring) | Professional › Finance, accounting & tax | 40.5 | 32.8 | −7.7 |
| DAYJOB: Accounting | Professional › Finance, accounting & tax | 35.2 | 27.6 | −7.6 |
| AA-3D v1.0 Elo | Software › Frontend & design | 1204 | 1259 | +7.6 |
| GDP.xlsx | Professional › General knowledge work | 29.1 | 22.3 | −6.8 |
| DAYJOB: Tax | Professional › Finance, accounting & tax | 10.8 | 4.8 | −6.0 |
| SimpleBench | Knowledge & reasoning › General knowledge | 75.9 | 69.9 | −6.0 |
| DAYJOB: Law (November 2026 release) | Professional › Legal | 26.8 | 32.7 | +5.9 |
| AA-Omniscience v1.1 Index | Knowledge & reasoning › General knowledge | 30.5 | 40.4 | +5.0 |
| Vibe Code Bench v1.1 | Software › Frontend & design | 92.4 | 87.6 | −4.8 |
| Medical Long Context Reasoning (MLCR-AA, October 2026 methodology) | Health & medicine | 41.7 | 37.2 | −4.4 |
| GDP.pdf (Artificial Analysis, All-pass) | Professional › General knowledge work | 25.8 | 30.0 | +4.2 |
| DualEntry Accounting AI Benchmark | Professional › Finance, accounting & tax | 80.2 | 84.2 | +4.0 |
| LMArena Agent Arena Bash Reliability | Tool & computer use › General agents | 11.9 | 4.3 | −3.8 |
| Dual-Arm Sim (task completion) | Tool & computer use › Robotics | 0.0 | 3.8 | +3.8 |
| AA-AnalystAgent (pass^5) | Professional › General knowledge work | 57.5 | 53.8 | −3.7 |
| VIALS-AA (pass@3) | Science › Life sciences | 70.8 | 74.5 | +3.7 |
| LMArena Agent Arena Praise vs Complaint | Tool & computer use › General agents | 30.7 | 23.3 | −3.7 |
| Terminal-Bench 4.0 (Artificial Analysis) | Software › Software engineering | 63.6 | 60.1 | −3.5 |
| LiveBench 2.0 (2026-11-06 release) | Composite indexes | 76.8 | 80.3 | +3.5 |
| VIALS-AA (average of 3) | Science › Life sciences | 65.4 | 68.9 | +3.5 |
| LMArena Agent Arena Confirmed Success | Tool & computer use › General agents | 17.6 | 10.9 | −3.3 |
| Artificial Analysis Economics Index | Composite indexes | 54.1 | 57.4 | +3.3 |
| RiemannBench (November 2026 rescoring) | Math | 76.8 | 80.0 | +3.2 |
| VIALS-AA (pass^3) | Science › Life sciences | 59.6 | 62.7 | +3.1 |
| CritPt v2 (agentic, avg@5) | Science › Physical sciences | 75.7 | 78.6 | +2.9 |
| Artificial Analysis Legal Index | Professional › Legal | 52.2 | 55.0 | +2.9 |
| Artificial Analysis Intelligence Index v5.0.1 | Composite indexes | 54.5 | 57.3 | +2.7 |
| FrontierCode 2.0 (Artificial Analysis) | Software › Software engineering | 41.7 | 44.4 | +2.7 |
| Artificial Analysis Engineering Index | Science › Engineering | 58.5 | 61.2 | +2.7 |
| GDP.pdf (November 2026 rescoring) | Professional › General knowledge work | 51.2 | 48.8 | −2.4 |
| ComplexConstraints (November 2026 rescoring) | Language & writing › Instruction following | 67.2 | 69.5 | +2.3 |
| ARC-AGI-2 | Knowledge & reasoning › Abstract reasoning | 83.9 | 85.8 | +1.9 |
| HLAB-Vals-1.1 | Professional › Legal | 10.4 | 12.1 | +1.7 |
| Excel Modeling Benchmark v1.1 | Professional › Finance, accounting & tax | 75.9 | 74.3 | −1.6 |
| Artificial Analysis Finance & Accounting Index | Professional › Finance, accounting & tax | 55.8 | 57.2 | +1.4 |
| Artificial Analysis Strategy & Ops Index | Professional › General knowledge work | 58.1 | 59.4 | +1.3 |
| Gray Swan IPI Attack Success (k=15) | Safety & alignment | 3.4 | 2.3 | +1.1 |
| ARC-AGI-1 Semi-Private | Knowledge & reasoning › Abstract reasoning | 96.5 | 97.5 | +1.0 |
| TxBench Preclinical Pharmacology V1 Clean Reruns | Science › Life sciences | 64.3 | 65.3 | +1.0 |
| Artificial Analysis Healthcare & Medical Index | Health & medicine | 54.2 | 53.3 | −0.9 |
| LMArena Agent Arena Steering Burden | Tool & computer use › General agents | 14.8 | 13.4 | −0.7 |
| AutomationBench-AA 1.0.6 | Tool & computer use › Customer service & operations | 71.8 | 71.3 | −0.5 |
| LMArena Agent Arena Tool Hallucination | Tool & computer use › Tool calling | 1.0 | 1.2 | +0.1 |
Data: Artificial Analysis, Surge AI, Arena and 16 othersatlas.village.dev