Atlas

Compare Models

GPT-6.1 Terra vs Muse Wonder Full

GPT-6.1 Terra leads by 0.6 capability points (≈95% interval −1.9 to +3.1): statistically tied · P(higher) ≈ 69%

Muse Wonder Full wins 31 of 44 shared benchmarks (13 lost)

Capability vs. relative cost
GPT-6.1 TerraMuse Wonder FullClosed 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-6.1 TerravsMuse Wonder Fullcapability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins13 – 31 · 44 shared

Comparing matching setups only (11 benchmarks excluded) · suites counted once
← GPT-6.1 TerraMuse Wonder Full →

GPT-6.1 TerravsMuse Wonder Fullwins by categoryData Nov 15, 2026Data: Artificial Analysis, Vals AI, Arena and 2 othersVillage LLM Atlasatlas.village.dev

Margins

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

Legal Research BenchLegal26.053.4+27.4
HLAB-Vals-1.1Legal1.723.3+21.7
Terminal-Bench-Science 0.1Science57.637.1−20.5
Tax Agent BenchFinance & Accounting55.075.0+20.0
Vending-Bench 2Games910712292+19.3
Vals Legal IndexIndexes58.877.9+19.1
AA-Briefcase v1.1 EloAgents14011729+18.0
GDPval-AA v2.1Professional Work14761857+16.8
Finance Agent v2.1Finance & Accounting56.067.6+11.6
Vals Finance IndexIndexes53.364.6+11.2
AutomationBench-AA 1.0.6Agents60.271.3+11.2
Code Migration v1.1 (Vals Index v2.2 subset)Code72.261.7−10.5
FrontierCode 1.1 Main (Artificial Analysis)Code36.044.9+8.9
Terminal-Bench 4.0Code49.558.1+8.6
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & Medicine30.622.2−8.3
Artificial Analysis Strategy & Ops IndexIndexes50.057.9+7.9
Artificial Analysis Intelligence Index v4.3.2Indexes48.155.8+7.7
SciCode 1.0.1 (Artificial Analysis)Code55.162.6+7.5
Humanity's Last Exam (Text-Only)General QA46.253.6+7.4
Artificial Analysis Healthcare & Medical IndexIndexes44.751.9+7.2
CritPt v2 (agentic, avg@5)Science81.475.1−6.3
Vals Index v2.2Indexes60.566.8+6.3
Artificial Analysis Finance & Accounting IndexIndexes49.055.2+6.2
LMArena Agent Arena Net Improvement (Steering Burden)Agents12.09.0−5.7
Artificial Analysis Economics IndexIndexes48.554.1+5.6
CritPt (avg@5)Science28.032.9+4.9
FrontierCode 2.0 (Artificial Analysis)Code42.337.8−4.6
Artificial Analysis Legal IndexIndexes46.450.7+4.3
GDP.pdf (Artificial Analysis, All-pass)Professional Work28.632.6+4.0
LMArena Agent Arena Steering BurdenAgents16.69.6−3.5
LMArena Agent Arena Bash ReliabilityAgents13.48.3−2.6
LMArena Agent Arena Praise vs ComplaintAgents20.015.4−2.3
Vibe Code Bench v1.1Code84.086.0+2.0
Excel Modeling Benchmark v1.1Finance & Accounting73.375.0+1.7
Artificial Analysis Intelligence Index v5.0.1Indexes51.350.1−1.2
Artificial Analysis Intelligence Index v5.0Indexes51.049.7−1.2
Humanity's Last Exam DiamondGeneral QA47.348.5+1.2
AA-LCR v1.1Search82.783.7+1.0
LMArena Agent Arena Confirmed SuccessAgents9.711.1+0.7
Artificial Analysis Progress Index v1Indexes345343−0.7
Artificial Analysis Engineering IndexIndexes54.755.3+0.5
GDPxProfessional Work8.59.0+0.5
LMArena Agent Arena Tool HallucinationAgents1.00.0−0.5
AA-Omniscience IndexIndexes29.830.1+0.2

GPT-6.1 TerravsMuse Wonder Fullnormalised score marginsData Nov 15, 2026Data: Artificial Analysis, Vals AI, Arena 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…