Atlas

Compare Models

GPT-5.6 Terra vs Muse Spark 1.3

Muse Spark 1.3 leads by 0.5 capability points (≈95% interval −1.5 to +2.5): statistically tied · P(higher) ≈ 68%

Wins 105 of 156 shared benchmarks (42 lost, 9 tied)

Capability vs. relative cost
GPT-5.6 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-5.6 TerravsMuse Spark 1.3capability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins42 – 105 · 156 shared, 9 tied

Comparing matching setups only (41 benchmarks excluded) · suites counted once
← GPT-5.6 TerraMuse Spark 1.3 →

GPT-5.6 TerravsMuse Spark 1.3wins by categoryData Nov 15, 2026Data: Vals AI, Artificial Analysis, Mercor and 33 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-5.6 Terra.

Harvey's Legal Agent Benchmark - Corporate GovernanceLegal0.060.0+60.0
Harvey's Legal Agent Benchmark - Data Privacy CybersecurityLegal0.060.0+60.0
BabaIsBench Stage 0: The IntroGames88.046.0−42.0
Harvey's Legal Agent Benchmark - Bankruptcy RestructuringLegal0.040.0+40.0
Harvey's Legal Agent Benchmark - Capital MarketsLegal0.040.0+40.0
Harvey's Legal Agent Benchmark - Emerging Companies Venture CapitalLegal0.040.0+40.0
Harvey's Legal Agent Benchmark - Healthcare Life SciencesLegal0.040.0+40.0
Harvey's Legal Agent Benchmark - ImmigrationLegal0.040.0+40.0
RiemannBenchMath66.428.0−38.4
IOI (Vals v2)Code87.656.6−31.1
OpenAI MRCR v2 8-needle - 512K-1MGeneral QA72.598.1+25.6
SWE-bench Verified Extended (Mercor held-out)Code78.754.0−24.7
FrontierMath Tier 4 (v2) - PrivateMath70.746.3−24.4
ChessBench CoherenceGames69.791.2+21.5
Gray Swan IPI Attack Success (k=15)Safety37.315.9+21.4
Harvey's Legal Agent Benchmark - Antitrust CompetitionLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Arbitration International Dispute ResolutionLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Banking FinanceLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Corporate M&ALegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Energy Natural ResourcesLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Funds Asset ManagementLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - International Trade SanctionsLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Real EstateLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Trusts Estates Private ClientLegal0.020.0+20.0
Olam Social Poker EloGames15111494−18.5
GeneBench-Pro Public Release SubsetScience8.026.0+18.0
WordleBenchGames91.073.0−18.0
ITBench-AAAgents51.033.2−17.8
HackerRank-BenchCode26.142.9+16.8
ChessBench EloGames9251201+16.4
ProofBench v1.1Math74.058.0−16.0
CadQueryEval (Corrected Geometry)Code68.084.0+16.0
Finance Agent v2 - General Qualitative AnalysisFinance & Accounting67.883.4+15.6
AA-Briefcase EloAgents13301589+15.1
GameBench 2Games53.568.5+15.0
Harvey HLAB — Criteria Pass RateLegal79.894.7+15.0
Humanity's Sixth SenseMultimodal22.737.4+14.7
Shroombench (mushroom identification, top-1)Multimodal35.049.0+14.0
Legal Research BenchLegal41.355.3+13.9
MMMU-Pro Extended (Mercor held-out)Multimodal47.433.8−13.6
APEX-AccountingFinance & Accounting40.353.5+13.2
DAYJOB: FinanceFinance & Accounting1.814.8+13.0
Harvey LAB (Mercor run, undisclosed ~213-task cohort)Legal2.315.0+12.7
DeepsecBench-AA (Artificial Analysis)Cybersecurity25.112.5−12.5
AA-Omniscience IndexIndexes0.125.0+12.5
Medical Long Context Reasoning (MLCR-AA)Health & Medicine31.743.3+11.7
FrontierMath Tiers 1-3 (v2) - PrivateMath86.074.4−11.6
Vibe Code Bench v1.1Code74.685.9+11.3
GBENCH One-shot Coding GScoreGames63.352.7−10.6
DeepSWE v1.1Code64.875.4+10.6
LMArena Agent Arena Net ImprovementAgents0.63.8+10.5
APEX-Agents 1.1Professional Work71.861.3−10.5
Finance Agent v2 - Financial ModelingFinance & Accounting23.033.5+10.5
Finance Agent v2 - PrecedentsFinance & Accounting24.835.2+10.4
𝜏³-BankingAgents40.250.5+10.3
Chess Puzzles (task version 2.0)Games47.037.0−10.0
Mystery Game PuzzlesGames35.025.0−10.0
GDPval-AA v2.1Professional Work14611681+9.7
Finance Agent v2 - Market AnalysisFinance & Accounting60.669.9+9.3
MedXpertQA MM Extended (Mercor held-out)Health & Medicine51.060.1+9.1
GDPval-AA v2Professional Work14791662+9.0
OpenAI MRCR v2 8-needle - 256K-512KGeneral QA89.698.5+8.9
GDPval Extended (Mercor held-out rubric score)Professional Work47.255.9+8.7
DistressBenchSafety76.484.9+8.4
APEX-SWECode31.139.1+8.0
LMArena Agent Arena Net Improvement (Steering Burden)Agents-1.82.2+7.6
TextQuests Game ProgressGames40.533.0−7.6
AA-LCR Extended (Mercor held-out)Search74.466.9−7.5
Humanity's Last ExamGeneral QA35.943.2+7.3
Tax Agent BenchFinance & Accounting65.272.4+7.2
Finance Agent v2 - All PassFinance & Accounting42.849.5+6.8
Extended NYT Connections 940 (Quadratic v1)Games78.485.1+6.7
Vision Arena OverallMultimodal12661291+6.4
ChartographyMultimodal34.027.6−6.4
Last Translation Benchmark (LTBv1-eval, blind)Chat & Writing12.218.4+6.3
NoCoT-Bench NCKI (kspine v3)General QA104113+6.2
DAYJOB: HealthcareHealth & Medicine1.67.8+6.2
ALE-BenchCode21541981−6.1
Artificial Analysis Intelligence Index v4.3.2Indexes42.148.1+6.0
Roboflow Vision Evals (September 2026 protocol)Multimodal73.879.8+6.0
AutomationBench 1.0.6 (Zapier)Agents14.820.7+5.9
Artificial Analysis Intelligence Index v4.3Indexes42.348.2+5.9
SRE Bench — Fully SolvedCode9.53.8−5.7
PCT Political Inconsistency (CAIS)Safety46.240.6+5.5
LMArena Agent Arena Confirmed SuccessAgents-7.23.7+5.5
Finance Agent v2 - General Quantitative AnalysisFinance & Accounting72.778.0+5.3
NoCoT-Bench NCRI 15.2General QA89.098.3+5.1
Vals Index v2.1Indexes53.158.2+5.1
LMArena Text Arena EloChat & Writing14681496+5.0
Vals Index v2Indexes59.664.5+4.9
Humanity's Last Exam DiamondGeneral QA28.833.7+4.9
Longform Creative WritingChat & Writing78.082.8+4.8
SciCode 1.0.1 (Artificial Analysis)Code55.059.7+4.7
Hedge-BenchFinance & Accounting13.89.1−4.7
Document ArenaProfessional Work14681473+4.7
Agents' Last Exam (ALE)Agents27.632.2+4.6
ChessBench AccuracyGames71.976.0+4.4
Terminal-Bench Science 0.1 (Vals)Science10.014.3+4.3
Artificial Analysis Intelligence Index v5.0.1Indexes32.336.6+4.3
AA-LCR v1.1Search79.083.0+4.0
CritPt (avg@5)Science30.026.0−4.0
Terminal-Bench 2.1Code88.084.3−3.7
BuseyBench SVGMultimodal6.15.7−3.7
APEX-1 (Mercor, mean score)Professional Work69.573.0+3.5
CVDP (517-problem set)Code50.954.3+3.4
DTBenchGeneral QA93.396.5+3.2
Finance Agent v2 - Disclosure AnalysisFinance & Accounting66.970.1+3.1
Finance Agent v2 - AdjustmentsFinance & Accounting49.752.8+3.1
Terminal-Bench-Science 0.1Science13.811.0−2.9
LMArena Agent Arena Bash RecoveryAgents0.96.1+2.6
Creative Writing v3 EloChat & Writing18561907+2.6
Epoch Capabilities Index (ECI)Indexes161158−2.3
Terminal-Bench 4.0 (Artificial Analysis)Code35.433.3−2.0
Eval Connections One-Shot with TrapsGames86.084.0−2.0
GDP.pdf (Artificial Analysis, All-pass)Professional Work24.626.6+2.0
ProgramBenchCode0.52.5+2.0
LMArena Agent Arena Praise vs ComplaintAgents-1.91.7+1.8
AutomationBench-AA 1.0.6Agents59.657.9−1.8
CritPt v2 (agentic, avg@5)Science69.768.0−1.7
LMArena Agent Arena Steering BurdenAgents-0.52.6+1.6
Terminal-Bench 4.0Code26.327.8+1.5
ObviousBenchGeneral QA97.999.3+1.4
Creative Writing v3 Rubric ScoreChat & Writing15.915.6−1.3
LMArena Agent Arena SteerabilityAgents4.11.6−1.3
Excel Modeling BenchmarkFinance & Accounting66.267.4+1.2
LMCAGeneral QA55.053.9−1.1
FrontierCode 2.0 (Artificial Analysis)Code23.923.1−0.8
CharXiv Extended (Mercor held-out)Multimodal81.181.8+0.7
SpeciEval (Pooled English Runs)Safety98.198.7+0.7
Conceptual Reasoning IndexIndexes65.766.2+0.7
SpeciEval (Refusal-Aware Retries, Historical)Safety98.198.8+0.7
CyberBench v1.1 OverallCybersecurity72.172.7+0.7
LMArena Agent Arena Bash ReliabilityAgents0.72.0+0.6
GPQA DiamondGeneral QA91.991.3−0.6
MedXpertQA MM (200-question set)Health & Medicine75.776.3+0.6
OTIS Mock AIME 2024-2025Math99.799.2−0.6
ACCoRDGeneral QA74.775.1+0.4
Code MigrationCode47.847.4−0.4
CursorBench 4.0Code41.341.6+0.3
LiveBench 2.0Agents75.074.8−0.2
Artificial Analysis Progress Index v1Indexes324324+0.2
CharXiv (descriptive and reasoning combined)Multimodal94.193.9−0.2
Finance Agent v2 - ComparablesFinance & Accounting47.047.2+0.1
Ruby LLM Benchmarks — Program FixerCode74.074.1+0.1
Finance Agent v2 - Earnings AnalysisFinance & Accounting77.377.2−0.1
LMArena Agent Arena Tool HallucinationAgents1.00.8−0.1
Financial Coaching BenchFinance & Accounting71.071.00.0
AIME 2025Math100.0100.00.0
Harvey's Legal Agent Benchmark - Employment LaborLegal0.00.00.0
Harvey's Legal Agent Benchmark - Environmental ESGLegal0.00.00.0
Harvey's Legal Agent Benchmark - InsuranceLegal0.00.00.0
Harvey's Legal Agent Benchmark - Intellectual PropertyLegal0.00.00.0
Harvey's Legal Agent Benchmark - Litigation Dispute ResolutionLegal0.00.00.0
Harvey's Legal Agent Benchmark - Structured Finance SecuritizationLegal20.020.00.0
Harvey's Legal Agent Benchmark - TaxLegal0.00.00.0
Harvey's Legal Agent Benchmark - White Collar Defense InvestigationsLegal0.00.00.0

GPT-5.6 TerravsMuse Spark 1.3normalised score marginsData Nov 15, 2026Data: Vals AI, Artificial Analysis, Mercor and 33 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…