Atlas

Compare Models

GPT-6.1 Terra vs Muse Spark 1.3

GPT-6.1 Terra leads by 4.1 capability points (≈95% interval +1.8 to +6.4): a clear lead over Muse Spark 1.3 · P(higher) ≈ 100%

Wins 42 of 58 shared benchmarks (16 lost)

Capability vs. relative cost
GPT-6.1 TerraMuse Spark 1.3Closed 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 Spark 1.3capability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins42 – 16 · 58 shared

Comparing matching setups only (17 benchmarks excluded) · suites counted once
← GPT-6.1 TerraMuse Spark 1.3 →

GPT-6.1 TerravsMuse Spark 1.3wins by categoryData Nov 15, 2026Data: Artificial Analysis, Arena, Epoch AI and 17 othersVillage LLM Atlasatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is Muse Spark 1.3 minus GPT-6.1 Terra.

VoxelBench Text RatingGames23131616−49.6
Mystery Game PuzzlesGames74.025.0−49.0
Terminal-Bench-Science 0.1Science57.611.0−46.7
FrontierMath Tier 4 (v2) - PrivateMath92.746.3−46.3
ChartographyMultimodal63.227.6−35.6
ALE-BenchCode28751981−31.7
Legal Research BenchLegal26.055.3+29.3
GBENCH One-shot Coding GScoreGames77.652.7−24.9
Terminal-Bench 4.0Code49.524.7−24.7
Terminal-Bench 4.0 (Artificial Analysis)Code55.633.3−22.2
WeirdML v3Science28.16.7−21.4
FrontierCode 2.0 (Artificial Analysis)Code42.323.1−19.2
DistressBenchSafety65.984.9+19.0
LMArena Agent Arena Net Improvement (Steering Burden)Agents12.02.2−18.5
APEX-Agents 1.1Professional Work79.361.3−18.0
TaxCalcBench TY25Finance & Accounting24.06.0−18.0
NoCoT-Bench NCRI 15.2General QA13098.3−17.7
Tax Agent BenchFinance & Accounting55.072.4+17.4
GameBench 2Games85.868.5−17.3
FrontierMath Tiers 1-3 (v2) - PrivateMath90.974.4−16.5
BuseyBench SVGMultimodal7.35.7−15.7
τ³-Banking (Mercor run, bm25_grep retrieval)Agents37.152.2+15.1
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & Medicine30.615.6−15.0
Artificial Analysis Intelligence Index v5.0.1Indexes51.336.6−14.7
Artificial Analysis Intelligence Index v5.0Indexes51.036.4−14.6
Humanity's Last Exam DiamondGeneral QA47.333.7−13.6
CritPt v2 (agentic, avg@5)Science81.468.0−13.4
Visa Agentic Commerce BenchmarkAgents51.164.4+13.3
Harvey LAB (Mercor run, undisclosed ~213-task cohort)Legal4.515.0+10.5
AA-Briefcase v1.1 EloAgents14011581+9.9
LMArena Agent Arena Praise vs ComplaintAgents20.01.7−9.2
GDPval-AA v2.1Professional Work14761681+9.0
APEX-1 (Mercor, mean score)Professional Work64.273.0+8.8
DAXBench v2Code93.685.1−8.5
Epoch Capabilities Index (ECI)Indexes168158−8.2
Artificial Analysis Progress Index v1Indexes345324−7.3
LMArena Agent Arena Steering BurdenAgents16.62.6−7.0
ChessBench CoherenceGames84.891.2+6.4
LMArena Agent Arena Bash ReliabilityAgents13.42.0−5.7
LiveBench 2.0Agents79.774.8−4.9
SciCode 1.0.1 (Artificial Analysis)Code55.159.7+4.6
ChessBench EloGames12511201−3.0
LMArena Agent Arena Confirmed SuccessAgents9.73.7−3.0
FrontierCode 1.1 Main (Artificial Analysis)Code41.438.6−2.7
ChessBench AccuracyGames78.376.0−2.5
Humanity's Last Exam (Text-Only)General QA46.248.7+2.5
AA-Omniscience IndexIndexes29.825.0−2.4
AutomationBench-AA 1.0.6Agents60.257.9−2.3
NoCoT-Bench NCKI (kspine v3)General QA110113+2.1
CritPt (avg@5)Science28.026.0−2.0
GDP.pdf (Artificial Analysis, All-pass)Professional Work28.626.6−2.0
Vibe Code Bench v1.1Code84.085.9+1.9
APEX-Accounting 1.1Finance & Accounting63.764.9+1.2
HiL-Bench (Human-in-Loop Benchmark)Agents68.066.9−1.1
OTIS Mock AIME 2024-2025Math100.099.2−0.8
AA-LCR v1.1Search82.783.0+0.3
LMArena Agent Arena Tool HallucinationAgents1.00.8−0.1
Artificial Analysis Intelligence Index v4.3.2Indexes48.148.10.0

GPT-6.1 TerravsMuse Spark 1.3normalised score marginsData Nov 15, 2026Data: Artificial Analysis, Arena, Epoch AI and 17 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…