Atlas

Compare Models

Claude Fable 5 vs GPT-6.1 Terra

GPT-6.1 Terra leads by 1.8 capability points (≈95% interval −0.7 to +4.3): statistically tied · P(higher) ≈ 92%

Wins 33 of 42 shared benchmarks (8 lost, 1 tied)

Capability vs. relative cost
Claude Fable 5GPT-6.1 TerraClosed 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 5vsGPT-6.1 Terracapability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins8 – 33 · 42 shared, 1 tied

Comparing matching setups only (18 benchmarks excluded) · suites counted once
← Claude Fable 5GPT-6.1 Terra →

Claude Fable 5vsGPT-6.1 Terrawins by categoryData Nov 15, 2026Data: Epoch AI, Arena, Artificial Analysis and 19 othersVillage LLM Atlasatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is GPT-6.1 Terra minus Claude Fable 5.

Furniture Assembly BenchmarkMultimodal35.879.2+43.3
Terminal-Bench-Science 0.1Science21.458.3+36.9
ChartographyMultimodal34.863.2+28.4
Vending-Bench 2Games45309107+27.7
Mystery Game PuzzlesGames52.074.0+22.0
VoxelBench Text RatingGames21012313+15.1
Tax Agent BenchFinance & Accounting69.855.0−14.8
ALE-BenchCode24632875+14.6
Terminal-Bench 4.0 (Artificial Analysis)Code42.455.6+13.1
NoCoT-Bench NCKI (kspine v3)General QA128110−12.8
ChessBench EloGames10461251+12.2
GBENCH One-shot Coding GScoreGames65.677.6+12.0
MMMU ProMultimodal72.784.7+12.0
HiL-Bench (Human-in-Loop Benchmark)Agents56.368.0+11.7
Agents' Last Exam Overall ScoreAgents48.760.0+11.3
GPQA DiamondGeneral QA83.394.3+11.0
GameBench 2Games75.285.8+10.6
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & Medicine20.630.6+10.0
LMArena Agent Arena Net Improvement (Steering Burden)Agents7.312.0+8.9
AA-Briefcase v1.1 EloAgents15361401−7.4
Vibe Code Bench v1.1Code90.484.0−6.4
NoCoT-Bench NCRI 15.2General QA119130+6.3
TaxCalcBench TY25Finance & Accounting30.024.0−6.0
Terminal-Bench 4.0Code44.550.3+5.8
Humanity's Last Exam (Text-Only)General QA51.546.2−5.3
APEX-Agents 1.1Professional Work74.579.3+4.8
LMArena Agent Arena Steering BurdenAgents7.916.6+4.4
LiveBench 2.0Agents83.679.7−4.0
FrontierMath Tiers 1-3 (v2) - PrivateMath87.090.9+3.9
Blueprint-Bench 2Multimodal0.40.4+3.5
LMArena Agent Arena Praise vs ComplaintAgents14.320.0+2.9
APEX-Accounting 1.1Finance & Accounting61.063.7+2.7
FrontierMath Tier 4 (v2) - PrivateMath90.292.7+2.5
LMArena Agent Arena Confirmed SuccessAgents5.19.7+2.3
ChessBench CoherenceGames82.684.8+2.2
APEX-1 (Mercor, mean score)Professional Work66.064.2−1.8
Humanity's Last Exam DiamondGeneral QA45.647.3+1.7
LMArena Agent Arena Bash ReliabilityAgents10.413.4+1.5
ChessBench AccuracyGames77.178.3+1.3
Prophet Arena Arena Score (Fixed Context)Forecasting84.985.2+0.3
LMArena Agent Arena Tool HallucinationAgents0.41.0+0.3
OTIS Mock AIME 2024-2025Math100.0100.00.0

Claude Fable 5vsGPT-6.1 Terranormalised score marginsData Nov 15, 2026Data: Epoch AI, Arena, Artificial Analysis and 19 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…