Atlas

Compare Models

GPT-6.2 Sol vs Kimi K3.1 Pro

GPT-6.2 Sol leads by 3.1 capability points (≈95% interval +0.4 to +5.7): a clear lead over Kimi K3.1 Pro · P(higher) ≈ 99%

Wins 16 of 27 shared benchmarks (8 lost, 3 tied)

Capability vs. relative cost
GPT-6.2 SolKimi K3.1 ProClosed weightsOpen weights

Each line runs through one model’s priced reasoning settings, cheapest to most expensive; grey points are other models’ priced settings in the same range. Results whose reasoning effort the source didn’t report are left out wherever a model has named effort levels. Capability and relative cost are the published estimates (relative cost is cost per task as a multiple of GPT-5 (high)); hover a setting, or use the arrow keys, to set it beside the other model’s same setting.

GPT-6.2 SolvsKimi K3.1 Procapability vs. relative costData Nov 17, 2026Village LLM Atlasatlas.village.dev

Benchmark wins16 – 8 · 27 shared, 3 tied

Comparing matching setups only (44 benchmarks excluded) · suites counted once
← GPT-6.2 SolKimi K3.1 Pro →

GPT-6.2 SolvsKimi K3.1 Prowins by categoryData Nov 17, 2026Data: Vals AI, Artificial Analysis, Moonshot AI and 2 othersVillage LLM Atlasatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is Kimi K3.1 Pro minus GPT-6.2 Sol.

Terminal-Bench-Science 0.1Science68.038.1−29.9
Terminal-Bench Science 0.1 (Vals)Science58.631.4−27.1
Code Migration v1.1 (Vals Index v2.2 subset)Code85.359.7−25.7
Humanity's Last Exam DiamondGeneral QA64.039.4−24.6
AA-3D v1.0 EloCode12591112−20.0
Legal Research Bench ImmigrationLegal36.418.2−18.2
Vals Science IndexIndexes57.844.4−13.4
Legal Research Bench Weighted PassLegal83.971.4−12.5
MysteryMechanismScience36.926.1−10.8
Legal Research Bench Administrative/RegulatoryLegal51.556.1+4.5
Excel Modeling Benchmark v1.1Finance & Accounting74.370.0−4.3
Legal Research Bench Civil LitigationLegal25.929.6+3.7
AA-Briefcase v1.1 EloAgents16121678+3.6
Terminal-Bench 4.0Code60.663.6+3.0
Vibe Code Bench v1.1Code87.684.9−2.7
Finance Agent v2.1Finance & Accounting58.961.5+2.6
Legal Research Bench Business/CommercialLegal30.227.9−2.3
BioMysteryBench (Vals Terminus 2)Science77.875.6−2.2
AutomationBench-AA 1.0.6Agents71.373.3+2.0
Legal Research Bench CriminalLegal29.828.1−1.8
Agents' Last Exam v1.1 ScoreAgents73.074.4+1.4
Tax Agent BenchFinance & Accounting62.963.9+1.0
Vals Finance IndexIndexes56.756.0−0.6
HLAB-Vals-1.1Legal12.111.7−0.4
Legal Research Bench Constitutional/Civil RightsLegal31.631.60.0
Legal Research Bench FamilyLegal13.613.60.0
Legal Research Bench HealthLegal73.373.30.0

GPT-6.2 SolvsKimi K3.1 Pronormalised score marginsData Nov 17, 2026Data: Vals AI, Artificial Analysis, Moonshot AI and 2 othersVillage LLM Atlasatlas.village.dev

Win counts use each benchmark’s declared score direction. Represented suite slices and the derived Capability score are excluded from headline totals. Missing or incompatible setups are excluded, never counted as losses.
Loading Atlas data…