Atlas

Compare Models

Claude Fable 5.1 vs Inkling 2 Small

Claude Fable 5.1 leads by 5.6 capability points (≈95% interval +3.3 to +7.9): a clear lead over Inkling 2 Small · P(higher) ≈ 100%

Wins 42 of 45 shared benchmarks (3 lost)

Capability vs. relative cost
Claude Fable 5.1Inkling 2 SmallClosed 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.

Claude Fable 5.1vsInkling 2 Smallcapability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins42 – 3 · 45 shared

Comparing matching setups only (59 benchmarks excluded) · suites counted once
← Claude Fable 5.1Inkling 2 Small →

Claude Fable 5.1vsInkling 2 Smallwins by categoryData Nov 15, 2026Data: Vals AI, Artificial Analysis, Surge AI and 13 othersVillage LLM Atlasatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is Inkling 2 Small minus Claude Fable 5.1.

ArXivMath 08/2026 (Native Harnesses)Math87.736.8−50.9
Terminal-Bench-Science 0.1Science52.65.7−46.9
HANDBOOK.md Agents v1.1Professional Work47.117.8−29.3
MysteryMechanismScience47.725.2−22.5
DAYJOB: Finance (October 2026 rescoring)Finance & Accounting36.514.3−22.2
CritPt v2 (agentic, avg@5)Science77.755.7−22.0
sudo L7 (October 2026 rescoring)Code32.214.1−18.1
Arena Code WebDev — ReactCode17881659−17.2
BrokenArXiv 08/2026 (Native Harnesses)Math79.863.1−16.7
AA-3D v1.0 EloCode11391027−16.2
APEX-Agents 1.1Professional Work77.160.9−16.2
DeepFinanceFinance & Accounting45.629.8−15.8
SWE-bench VerifiedCode96.682.2−14.4
SWE Atlas v1.1 - OverallCode74.161.9−12.2
Excel Modeling Benchmark v1.1 LBOFinance & Accounting68.055.9−12.1
Tax Agent BenchFinance & Accounting77.965.8−12.1
FrontierCode 2.0 - Pass RateCode37.625.5−12.1
FrontierCode 2.0 - ScoreCode34.022.6−11.5
SciCode 1.0.1 (Artificial Analysis)Code63.152.3−10.8
SWE-bench MultilingualCode91.181.2−9.9
ChartographyMultimodal46.236.7−9.5
Excel Modeling Benchmark v1.1 DCFFinance & Accounting71.562.2−9.3
Excel Modeling Benchmark v1.1 TemplateFinance & Accounting80.871.6−9.2
Excel Modeling Benchmark v1.1 CompsFinance & Accounting82.073.1−8.9
ARC-AGI-2General QA90.081.7−8.3
AA-Briefcase v1.1 EloAgents16711527−7.9
Humanity's Sixth SenseMultimodal40.833.3−7.5
Excel Modeling Benchmark v1.1 M&AFinance & Accounting80.573.7−6.9
SWE-Marathon v1.1Code45.638.8−6.8
AA-AnalystAgent (pass^5)Professional Work57.551.3−6.2
GDPval-AA v2.1Professional Work17651630−5.9
EnterpriseBench: CoreCraftProfessional Work77.471.8−5.6
Excel Modeling Benchmark v1.1 ScratchFinance & Accounting72.767.1−5.6
Excel Modeling Benchmark v1.1 OperatingFinance & Accounting75.270.0−5.2
Excel Modeling Benchmark v1.1 DataroomFinance & Accounting88.583.4−5.1
DeepSWE v1.1Code67.372.1+4.8
APEX-Accounting 1.1Finance & Accounting66.762.0−4.7
ATLAS-Finance v1.1Finance & Accounting10.36.0−4.3
Excel Modeling Benchmark v1.1 MiscFinance & Accounting71.468.3−3.1
Arena Code WebDev — HTMLCode16811660−3.1
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & Medicine18.320.6+2.2
GPQA DiamondGeneral QA91.493.0+1.6
ARC-AGI-1 Semi-PrivateGeneral QA97.596.5−1.0
ComplexConstraintsChat & Writing45.144.5−0.6
Terminal-Bench 2.1Code81.681.3−0.3

Claude Fable 5.1vsInkling 2 Smallnormalised score marginsData Nov 15, 2026Data: Vals AI, Artificial Analysis, Surge AI and 13 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…