Gemini 4 Neon vs GPT-6.2 Sol
Gemini 4 Neon leads by 2.0 capability points (≈95% interval −0.9 to +4.9): statistically tied · P(higher) ≈ 91%
Wins 52 of 59 shared benchmarks (7 lost)
Capability vs. relative cost
Gemini 4 NeonGPT-6.2 SolClosed weightsOpen weights
Benchmark wins52 – 7 · 59 shared
Comparing matching setups only (31 benchmarks excluded) · suites counted once
Area← Gemini 4 NeonGPT-6.2 Sol →
Data: Artificial Analysis, Vals AI, Surge AI and 7 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 Neon.
| EnterpriseBench: CoreCraft | Tool & computer use › Customer service & operations | 77.4 | 36.4 | −41.0 |
| DAYJOB: Healthcare (October 2026 rescoring) | Health & medicine | 47.2 | 17.2 | −30.0 |
| SimpleBench | Knowledge & reasoning › General knowledge | 94.4 | 69.9 | −24.5 |
| DAYJOB: Accounting | Professional › Finance, accounting & tax | 49.6 | 27.6 | −22.0 |
| GDP.xlsx | Professional › General knowledge work | 43.1 | 22.3 | −20.8 |
| Legal Research Bench | Professional › Legal | 56.3 | 37.0 | −19.2 |
| Humanity's Sixth Sense | Perception & media › Spatial & visual reasoning | 75.9 | 56.8 | −19.0 |
| HANDBOOK.md Agents v1.1 | Tool & computer use › Customer service & operations | 52.3 | 34.5 | −17.8 |
| DAYJOB: Finance (October 2026 rescoring) | Professional › Finance, accounting & tax | 50.5 | 32.8 | −17.7 |
| Tax Agent Bench | Professional › Finance, accounting & tax | 79.8 | 62.9 | −16.9 |
| GraphWalks BFS: 256K to 1M Aggregate | Knowledge & reasoning › Reading & long context | 95.2 | 80.1 | −15.1 |
| AutomationBench 1.0.6 (Zapier) | Tool & computer use › Customer service & operations | 56.3 | 41.3 | −15.1 |
| LMArena Agent Arena Net Improvement (Steering Burden) | Tool & computer use › General agents | 18.4 | 11.3 | −13.8 |
| Vals Legal Index | Professional › Legal | 78.4 | 66.7 | −11.7 |
| Vals Finance Index | Professional › Finance, accounting & tax | 68.3 | 56.7 | −11.7 |
| LMArena Agent Arena Praise vs Complaint | Tool & computer use › General agents | 44.5 | 23.3 | −10.6 |
| Finance Agent v2.1 | Professional › Finance, accounting & tax | 68.6 | 58.9 | −9.6 |
| Code Migration v1.1 (Vals Index v2.2 subset) | Software › Software engineering | 75.8 | 85.3 | +9.6 |
| Artificial Analysis Legal Index | Professional › Legal | 63.8 | 55.0 | −8.7 |
| MysteryMechanism | Math | 45.0 | 36.9 | −8.1 |
| AutomationBench-AA 1.0.6 | Tool & computer use › Customer service & operations | 79.3 | 71.3 | −8.1 |
| DAYJOB: Tax | Professional › Finance, accounting & tax | 12.8 | 4.8 | −8.0 |
| Artificial Analysis Finance & Accounting Index | Professional › Finance, accounting & tax | 65.1 | 57.2 | −7.8 |
| HLAB-Vals-1.1 | Professional › Legal | 19.6 | 12.1 | −7.5 |
| Humanity's Last Exam Diamond | Knowledge & reasoning › General knowledge | 71.3 | 64.0 | −7.3 |
| AA-Omniscience v1.1 Index | Knowledge & reasoning › General knowledge | 54.8 | 40.4 | −7.2 |
| LiveBench 2.0 (2026-11-06 release) | Composite indexes | 87.4 | 80.3 | −7.1 |
| Artificial Analysis Healthcare & Medical Index | Health & medicine | 60.1 | 53.3 | −6.8 |
| AA-Briefcase v1.1 Elo | Professional › General knowledge work | 1699 | 1580 | −6.5 |
| DAYJOB: Law (November 2026 release) | Professional › Legal | 26.2 | 32.7 | +6.5 |
| Vibe Code Bench v1.1 | Software › Frontend & design | 93.7 | 87.6 | −6.1 |
| Artificial Analysis Engineering Index | Science › Engineering | 66.8 | 61.2 | −5.7 |
| Medical Long Context Reasoning (MLCR-AA, October 2026 methodology) | Health & medicine | 31.7 | 37.2 | +5.6 |
| LMArena Agent Arena Bash Reliability | Tool & computer use › General agents | -6.0 | 4.3 | +5.1 |
| Vals Science Index | Science › Multidisciplinary | 62.6 | 57.8 | −4.9 |
| Vals Index v2.2 | Composite indexes | 70.9 | 66.1 | −4.8 |
| GDP.pdf (November 2026 rescoring) | Professional › General knowledge work | 53.4 | 48.8 | −4.6 |
| Artificial Analysis Economics Index | Composite indexes | 61.9 | 57.4 | −4.6 |
| Artificial Analysis Strategy & Ops Index | Professional › General knowledge work | 63.8 | 59.4 | −4.4 |
| Terminal-Bench Science 0.1 (Vals) | Science › Multidisciplinary | 62.9 | 58.6 | −4.3 |
| LMArena Agent Arena Confirmed Success | Tool & computer use › General agents | 19.2 | 10.9 | −4.1 |
| Artificial Analysis Intelligence Index v5.0.1 | Composite indexes | 61.4 | 57.3 | −4.1 |
| RiemannBench (November 2026 rescoring) | Math | 84.0 | 80.0 | −4.0 |
| LMArena Agent Arena Steering Burden | Tool & computer use › General agents | 21.2 | 13.4 | −3.9 |
| ComplexConstraints (November 2026 rescoring) | Language & writing › Instruction following | 72.9 | 69.5 | −3.4 |
| GDP.pdf (Artificial Analysis, All-pass) | Professional › General knowledge work | 33.4 | 30.0 | −3.4 |
| Terminal-Bench-Science 0.1 | Science › Multidisciplinary | 70.0 | 66.7 | −3.3 |
| Terminal-Bench 4.0 (Artificial Analysis) | Software › Software engineering | 61.6 | 58.6 | −3.0 |
| Agents' Last Exam v1.1 Score | Professional › General knowledge work | 75.8 | 73.0 | −2.8 |
| BioMysteryBench (Vals Terminus 2) | Science › Life sciences | 80.0 | 77.8 | −2.2 |
| sudo L7 (October 2026 rescoring) | Software › Software engineering | 38.3 | 36.1 | −2.2 |
| Excel Modeling Benchmark v1.1 | Professional › Finance, accounting & tax | 76.5 | 74.3 | −2.2 |
| FrontierCode 2.0 (Artificial Analysis) | Software › Software engineering | 44.4 | 42.8 | −1.7 |
| Chartography | Perception & media › Documents & charts | 73.5 | 75.1 | +1.6 |
| LVBench | Perception & media › Video understanding | 89.0 | 87.6 | −1.4 |
| SWE-sweep | Software › Software engineering | 12.8 | 11.6 | −1.2 |
| Terminal-Bench 4.0 | Software › Software engineering | 61.6 | 60.6 | −1.0 |
| CritPt v2 (agentic, avg@5) | Science › Physical sciences | 78.0 | 78.3 | +0.3 |
| LMArena Agent Arena Tool Hallucination | Tool & computer use › Tool calling | 1.0 | 1.2 | +0.1 |
Data: Artificial Analysis, Vals AI, Surge AI and 7 othersatlas.village.dev