Gemini 4 Argon vs GPT-6.2 Sol
GPT-6.2 Sol leads by 1.3 capability points (≈95% interval −1.3 to +3.9): statistically tied · P(higher) ≈ 84%
Wins 40 of 75 shared benchmarks (35 lost)
Capability vs. relative cost
Gemini 4 ArgonGPT-6.2 SolClosed weightsOpen weights
Benchmark wins35 – 40 · 75 shared
Comparing matching setups only (36 benchmarks excluded) · suites counted once
Area← Gemini 4 ArgonGPT-6.2 Sol →
Data: Artificial Analysis, Vals AI, Surge AI and 15 othersatlas.village.dev
Margins
Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is GPT-6.2 Sol minus Gemini 4 Argon.
| AA-PersonalityBench Slop index | Language & writing › Style & personality | 67.6 | 15.4 | −52.8 |
| AA-PersonalityBench Informativeness index | Language & writing › Style & personality | 82.9 | 37.7 | −45.6 |
| KradleBench | Games | 28.6 | 67.9 | +39.3 |
| AA-PersonalityBench Transparency index | Language & writing › Style & personality | 21.9 | 56.9 | +35.3 |
| AA-PersonalityBench Expressiveness index | Language & writing › Style & personality | 84.8 | 55.3 | −29.8 |
| DAYJOB: Law (November 2026 release) | Professional › Legal | 5.9 | 32.7 | +26.8 |
| Medical Long Context Reasoning (MLCR-AA, October 2026 methodology) | Health & medicine | 13.9 | 37.2 | +23.3 |
| SimpleBench | Knowledge & reasoning › General knowledge | 91.7 | 69.9 | −21.8 |
| sudo L7 (October 2026 rescoring) | Software › Software engineering | 16.1 | 36.1 | +20.0 |
| AA-PersonalityBench Warmth index | Language & writing › Style & personality | 79.1 | 60.1 | −19.2 |
| Legal Research Bench | Professional › Legal | 54.8 | 37.0 | −17.8 |
| Ramp SWE-Bench v2 (Bash harness) | Software › Software engineering | 54.8 | 72.4 | +17.6 |
| FrontierCode 2.0 (Artificial Analysis) | Software › Software engineering | 25.0 | 41.4 | +16.4 |
| Tax Agent Bench | Professional › Finance, accounting & tax | 79.0 | 62.9 | −16.1 |
| GDP.xlsx | Professional › General knowledge work | 38.3 | 22.3 | −16.0 |
| AA-PersonalityBench Conviction index | Language & writing › Style & personality | 63.8 | 79.2 | +15.6 |
| AA-3D v1.0 Elo | Software › Frontend & design | 1147 | 1259 | +15.3 |
| HLAB-Vals-1.1 | Professional › Legal | 27.1 | 12.1 | −15.0 |
| Terminal-Bench Science 0.1 (Vals) | Science › Multidisciplinary | 44.3 | 58.6 | +14.3 |
| AA-PersonalityBench Imagination index | Language & writing › Style & personality | 46.7 | 60.1 | +13.5 |
| Terminal-Bench-Science 0.1 | Science › Multidisciplinary | 48.6 | 60.5 | +11.9 |
| Code Migration v1.1 (Vals Index v2.2 subset) | Software › Software engineering | 74.4 | 85.3 | +10.9 |
| AA-PersonalityBench Organization index | Language & writing › Style & personality | 56.2 | 45.7 | −10.6 |
| Vals Legal Index | Professional › Legal | 76.8 | 66.7 | −10.1 |
| DAYJOB: Accounting | Professional › Finance, accounting & tax | 18.0 | 27.6 | +9.6 |
| SWE Atlas v1.1 - Overall | Software › Software engineering | 70.1 | 60.5 | −9.6 |
| Vals Finance Index | Professional › Finance, accounting & tax | 66.0 | 56.7 | −9.3 |
| AutomationBench 1.0.6 (Zapier) | Tool & computer use › Customer service & operations | 50.1 | 41.3 | −8.8 |
| Finance Agent v2.1 | Professional › Finance, accounting & tax | 67.7 | 58.9 | −8.7 |
| MysteryMechanism | Math | 45.5 | 36.9 | −8.6 |
| Dual-Arm Sim (task completion) | Tool & computer use › Robotics | 11.3 | 3.8 | −7.5 |
| ARC-AGI-3 (Semi-Private) | Games | 97.3 | 90.1 | −7.2 |
| DualEntry Accounting AI Benchmark | Professional › Finance, accounting & tax | 91.1 | 84.2 | −6.9 |
| Artificial Analysis Intelligence Index v5.0.1 | Composite indexes | 50.4 | 57.3 | +6.8 |
| MultiChallenge | Language & writing › Conversation | 78.7 | 72.0 | −6.7 |
| DAYJOB: Healthcare (October 2026 rescoring) | Health & medicine | 23.6 | 17.2 | −6.4 |
| AutomationBench-AA 1.0.6 | Tool & computer use › Customer service & operations | 77.5 | 71.3 | −6.2 |
| Professional Reasoning Benchmark - Legal | Professional › Legal | 49.4 | 43.3 | −6.1 |
| RiemannBench (November 2026 rescoring) | Math | 74.4 | 80.0 | +5.6 |
| ARC-AGI-2 | Knowledge & reasoning › Abstract reasoning | 91.3 | 85.8 | −5.4 |
| LMArena Agent Arena Bash Reliability | Tool & computer use › General agents | -6.0 | 4.3 | +5.1 |
| CritPt v2 (agentic, avg@5) | Science › Physical sciences | 72.9 | 77.7 | +4.9 |
| GDP.pdf (November 2026 rescoring) | Professional › General knowledge work | 53.6 | 48.8 | −4.8 |
| GDP.pdf (Artificial Analysis, All-pass) | Professional › General knowledge work | 25.6 | 30.0 | +4.4 |
| GraphWalks BFS: 256K to 1M Aggregate | Knowledge & reasoning › Reading & long context | 84.2 | 80.1 | −4.1 |
| LVBench | Perception & media › Video understanding | 91.7 | 87.6 | −4.1 |
| Agents' Last Exam v1.1 Score | Professional › General knowledge work | 69.0 | 73.0 | +4.0 |
| AA-AnalystAgent (pass^5) | Professional › General knowledge work | 48.8 | 52.5 | +3.8 |
| Chartography | Perception & media › Documents & charts | 71.6 | 75.1 | +3.5 |
| Humanity's Last Exam Diamond | Knowledge & reasoning › General knowledge | 53.2 | 56.5 | +3.3 |
| DrugDiscoveryBench | Science › Life sciences | 65.4 | 62.2 | −3.2 |
| VIALS-AA (pass^3) | Science › Life sciences | 59.6 | 62.7 | +3.1 |
| AA-Briefcase v1.1 Elo | Professional › General knowledge work | 1485 | 1541 | +3.1 |
| Terminal-Bench 4.0 | Software › Software engineering | 57.6 | 60.6 | +3.0 |
| AA-Omniscience v1.1 Index | Knowledge & reasoning › General knowledge | 46.1 | 40.4 | −2.9 |
| VIALS-AA (average of 3) | Science › Life sciences | 66.3 | 68.9 | +2.7 |
| LiveBench 2.0 (2026-11-06 release) | Composite indexes | 77.7 | 80.3 | +2.6 |
| Blueprint-Bench 2 | Perception & media › Spatial & visual reasoning | 0.5 | 0.6 | +2.6 |
| Vals Science Index | Science › Multidisciplinary | 55.4 | 57.8 | +2.4 |
| DAYJOB: Finance (October 2026 rescoring) | Professional › Finance, accounting & tax | 30.5 | 32.8 | +2.3 |
| HANDBOOK.md Agents v1.1 | Tool & computer use › Customer service & operations | 36.6 | 34.5 | −2.1 |
| LMArena Agent Arena Net Improvement (Steering Burden) | Tool & computer use › General agents | 10.3 | 11.3 | +2.0 |
| VIALS-AA (pass@3) | Science › Life sciences | 72.7 | 74.5 | +1.9 |
| DAYJOB: Tax | Professional › Finance, accounting & tax | 3.2 | 4.8 | +1.6 |
| BioMysteryBench (Vals Terminus 2) | Science › Life sciences | 76.3 | 77.8 | +1.5 |
| Humanity's Sixth Sense | Perception & media › Spatial & visual reasoning | 55.4 | 56.8 | +1.4 |
| Terminal-Bench 4.0 (Artificial Analysis) | Software › Software engineering | 55.1 | 56.1 | +1.0 |
| ARC-AGI-1 Semi-Private | Knowledge & reasoning › Abstract reasoning | 98.5 | 97.5 | −1.0 |
| LMArena Agent Arena Confirmed Success | Tool & computer use › General agents | 12.6 | 10.9 | −0.9 |
| ComplexConstraints (November 2026 rescoring) | Language & writing › Instruction following | 70.3 | 69.5 | −0.8 |
| TxBench Preclinical Pharmacology V1 Clean Reruns | Science › Life sciences | 64.7 | 65.3 | +0.7 |
| LMArena Agent Arena Steering Burden | Tool & computer use › General agents | 12.2 | 13.4 | +0.6 |
| Excel Modeling Benchmark v1.1 | Professional › Finance, accounting & tax | 74.8 | 74.3 | −0.5 |
| LMArena Agent Arena Praise vs Complaint | Tool & computer use › General agents | 23.8 | 23.3 | −0.2 |
| LMArena Agent Arena Tool Hallucination | Tool & computer use › Tool calling | 0.8 | 1.2 | +0.2 |
Data: Artificial Analysis, Vals AI, Surge AI and 15 othersatlas.village.dev