Atlas

Compare Models

Inkling 2 Small vs Kimi K3.1 Pro

Kimi K3.1 Pro leads by 3.7 capability points (≈95% interval +1.5 to +5.9): a clear lead over Inkling 2 Small · P(higher) ≈ 100%

Inkling 2 Small wins 42 of 88 shared benchmarks (40 lost, 6 tied)

Capability vs. relative cost
Inkling 2 SmallKimi 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.

Inkling 2 SmallvsKimi K3.1 Procapability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins42 – 40 · 88 shared, 6 tied

Comparing matching setups only (85 benchmarks excluded) · suites counted once
← Inkling 2 SmallKimi K3.1 Pro →

Inkling 2 SmallvsKimi K3.1 Prowins by categoryData Nov 15, 2026Data: Vals AI, MathArena, Moonshot AI and 9 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 Inkling 2 Small.

Terminal-Bench 4.0Code12.663.6+51.0
HLAB-Vals-1.1 - Antitrust CompetitionLegal40.00.0−40.0
HLAB-Vals-1.1 - Data Privacy CybersecurityLegal20.060.0+40.0
HLAB-Vals-1.1 - Energy Natural ResourcesLegal50.010.0−40.0
ArXivMath 09/2026 (Native Harnesses)Math31.164.2+33.0
Terminal-Bench-Science 0.1Science5.738.1+32.4
HLAB-Vals-1.1 - Funds Asset ManagementLegal30.00.0−30.0
HLAB-Vals-1.1 - Healthcare Life SciencesLegal30.00.0−30.0
HLAB-Vals-1.1 - ImmigrationLegal30.00.0−30.0
Terminal-Bench Science 0.1 (Vals)Science2.931.4+28.6
ArXivMath 08/2026 (Native Harnesses)Math36.864.9+28.1
SRE Bench — Fully SolvedCode1.927.1+25.2
Legal Research Bench Civil LitigationLegal51.929.6−22.2
Code Migration v1.1 (Vals Index v2.2 subset)Code39.559.7+20.2
HLAB-Vals-1.1 - Arbitration International Dispute ResolutionLegal30.010.0−20.0
HLAB-Vals-1.1 - Corporate GovernanceLegal30.010.0−20.0
HLAB-Vals-1.1 - Corporate M&ALegal20.00.0−20.0
HLAB-Vals-1.1 - Litigation Dispute ResolutionLegal40.020.0−20.0
HLAB-Vals-1.1 - TaxLegal20.00.0−20.0
Legal Research Bench Business/CommercialLegal46.527.9−18.6
ArXivMath 10/2026 (Native Harnesses)Math27.745.7+18.1
Legal Research Bench Weighted PassLegal87.971.4−16.4
Finance Agent v2.1 - PrecedentsFinance & Accounting47.131.7−15.4
APEX-Agents 1.1Professional Work60.975.9+15.0
Vals Science IndexIndexes32.744.4+11.7
Finance Agent v2.1 - Earnings AnalysisFinance & Accounting88.978.0−10.9
Humanity's Last Exam DiamondGeneral QA28.739.4+10.7
Finance Agent v2.1 - Disclosure AnalysisFinance & Accounting80.870.1−10.7
Finance Agent v2.1 - AdjustmentsFinance & Accounting67.457.4−10.0
HLAB-Vals-1.1 - Banking FinanceLegal10.00.0−10.0
HLAB-Vals-1.1 - Capital MarketsLegal20.030.0+10.0
HLAB-Vals-1.1 - Emerging Companies Venture CapitalLegal20.010.0−10.0
HLAB-Vals-1.1 - Intellectual PropertyLegal20.010.0−10.0
HLAB-Vals-1.1 - International Trade SanctionsLegal20.010.0−10.0
HLAB-Vals-1.1 - Real EstateLegal30.040.0+10.0
HLAB-Vals-1.1 - Trusts Estates Private ClientLegal20.030.0+10.0
GDPval-AA v2.1Professional Work16301846+9.5
Legal Research Bench FamilyLegal22.713.6−9.1
Legal Research Bench ImmigrationLegal27.318.2−9.1
Legal Research Bench CriminalLegal36.828.1−8.8
AA-Briefcase v1.1 EloAgents15271678+8.3
Finance Agent v2.1 - All PassFinance & Accounting56.648.4−8.2
Chartography (with tools)Multimodal82.190.2+8.1
Legal Research Bench Constitutional/Civil RightsLegal39.531.6−7.9
Finance Agent v2.1 - General Qualitative AnalysisFinance & Accounting78.070.8−7.3
Excel Modeling Benchmark v1.1 CompsFinance & Accounting73.167.4−5.8
EnterpriseBench: CoreCraftProfessional Work71.866.1−5.7
BioMysteryBench (Vals Terminus 2)Science70.075.6+5.6
Arena Code WebDev — ReactCode16591700+5.6
BrokenArXiv 10/2026 (Native Harnesses)Math63.358.0−5.2
DeepFinanceFinance & Accounting29.834.6+4.8
Legal Research Bench Administrative/RegulatoryLegal60.656.1−4.5
FrontierMemory v1.1Agents60.055.6−4.4
Excel Modeling Benchmark v1.1 M&AFinance & Accounting73.778.1+4.4
Arena Code WebDev — HTMLCode16601689+4.1
Excel Modeling Benchmark v1.1 DataroomFinance & Accounting83.487.6+4.1
Finance Agent v2.1 - Financial ModelingFinance & Accounting42.446.0+3.6
PerceptionBenchMultimodal64.768.3+3.6
FrontierCode 2.0 - Pass RateCode25.522.0−3.5
CharXiv ReasoningMultimodal85.689.0+3.4
DeepSWE v1.1Code72.175.5+3.4
Vibe Code Bench v1.1Code81.584.9+3.3
Legal Research Bench HealthLegal70.073.3+3.3
Finance Agent v2.1 - General Quantitative AnalysisFinance & Accounting79.276.1−3.1
DAYJOB: Finance (October 2026 rescoring)Finance & Accounting14.317.0+2.7
FrontierCode 2.0 - ScoreCode22.619.9−2.7
sudo L7 (October 2026 rescoring)Code14.116.7+2.6
Excel Modeling Benchmark v1.1 DCFFinance & Accounting62.264.4+2.2
Tax Agent BenchFinance & Accounting65.863.9−1.9
BabyVision with PythonMultimodal93.695.4+1.8
HLAB-Vals-1.1 - Criteria Pass RateLegal93.792.5−1.2
BrokenArXiv 08/2026 (Native Harnesses)Math63.164.3+1.2
Finance Agent v2.1 - ComparablesFinance & Accounting65.063.9−1.1
Excel Modeling Benchmark v1.1 ScratchFinance & Accounting67.168.1+1.0
MysteryMechanismScience25.226.1+0.9
Excel Modeling Benchmark v1.1 LBOFinance & Accounting55.955.2−0.7
Excel Modeling Benchmark v1.1 OperatingFinance & Accounting70.069.3−0.6
APEX-Accounting 1.1Finance & Accounting62.062.4+0.4
Finance Agent v2.1 - Market AnalysisFinance & Accounting59.059.4+0.4
Excel Modeling Benchmark v1.1 TemplateFinance & Accounting71.671.3−0.3
BrokenArXiv 09/2026 (Native Harnesses)Math63.263.5+0.3
Excel Modeling Benchmark v1.1 MiscFinance & Accounting68.368.1−0.1
HLAB-Vals-1.1 - Bankruptcy RestructuringLegal20.020.00.0
HLAB-Vals-1.1 - Employment LaborLegal0.00.00.0
HLAB-Vals-1.1 - Environmental ESGLegal0.00.00.0
HLAB-Vals-1.1 - InsuranceLegal0.00.00.0
HLAB-Vals-1.1 - Structured Finance SecuritizationLegal20.020.00.0
HLAB-Vals-1.1 - White Collar Defense InvestigationsLegal0.00.00.0

Inkling 2 SmallvsKimi K3.1 Pronormalised score marginsData Nov 15, 2026Data: Vals AI, MathArena, Moonshot AI and 9 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…