Atlas

Compare Models

GPT-6.1 Sol vs Muse Spark 1.3

GPT-6.1 Sol leads by 5.1 capability points (≈95% interval +2.8 to +7.4): a clear lead over Muse Spark 1.3 · P(higher) ≈ 100%

Wins 119 of 169 shared benchmarks (48 lost, 2 tied)

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

Benchmark wins119 – 48 · 169 shared, 2 tied

Comparing matching setups only (75 benchmarks excluded) · suites counted once
← GPT-6.1 SolMuse Spark 1.3 →

GPT-6.1 SolvsMuse Spark 1.3wins by categoryData Nov 15, 2026Data: Vals AI, Artificial Analysis, Mercor and 41 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 Sol.

AA-PersonalityBench Informativeness indexChat & Writing27.7100+73.1
VoxelBench Text RatingGames25501616−66.5
AA-PersonalityBench Slop indexChat & Writing14.375.3+61.5
Mystery Game PuzzlesGames80.025.0−55.0
FrontierMath Tier 4 (v2) - PrivateMath100.046.3−53.7
BrokenArXiv 05/2026Math90.036.5−53.5
Olam Social Poker EloGames15431494−53.3
Design Creativity BenchCode63.747.2−52.5
AA-PersonalityBench Transparency indexChat & Writing92.441.0−51.9
Terminal-Bench-Science 0.1Science58.111.0−47.1
SRE Bench — Fully SolvedCode50.83.8−46.9
BrokenArXiv 06/2026Math100.053.2−46.8
CheatBenchSafety42.886.5−43.7
Terminal-Bench Science 0.1 (Vals)Science52.910.0−42.9
ProofBench v1.1Math99.058.0−41.0
IOI (Vals v2)Code96.956.6−40.3
TextQuests Game ProgressGames71.433.0−38.5
SAFE-Teen 106 worst branch: Relationships (no prompt)Safety19.156.2−37.1
AA-PersonalityBench Imagination indexChat & Writing61.998.1+36.5
ALE-BenchCode29901981−35.7
CyberBench v1.1 OverallCybersecurity39.372.7+33.5
TaxCalcBench TY25Finance & Accounting38.06.0−32.0
WeirdML v3Science38.36.7−31.6
Chess Puzzles (task version 2.0)Games68.037.0−31.0
AA-PersonalityBench Warmth indexChat & Writing65.796.2+30.8
Terminal-Bench 4.0Code55.124.7−30.3
Decision-Bench v1.1Professional Work49.119.8−29.2
Code2Video-AA EloCode13691019−28.5
NoCoT-Bench NCRI 15.2General QA14998.3−27.9
MMMU-Pro Extended (Mercor held-out)Multimodal61.333.8−27.5
ATLAS (Exa, 200-task subset)Search69.542.7−26.8
GBENCH One-shot Coding GScoreGames79.052.7−26.3
RiemannBench (November 2026 rescoring)Math79.253.6−25.6
Humanity's Last Exam DiamondGeneral QA58.833.7−25.1
SAFE-Teen 106: Relationships (no prompt)Safety10.234.4−24.1
HoneyBench v0.2 (hack rate)Safety59.583.6−24.1
GameBench 2Games92.068.5−23.5
AA-PersonalityBench Conviction indexChat & Writing54.377.2+23.1
Terminal-Bench 4.0 (Artificial Analysis)Code56.133.3−22.7
BuseyBench SVGMultimodal8.05.7−22.4
ArXivMath 05/2026Math95.073.3−21.7
ChessBench EloGames15621201−21.5
AA-PersonalityBench Expressiveness indexChat & Writing50.571.4+21.2
sudo L7 (October 2026 rescoring)Code33.912.8−21.1
LMArena Agent Arena Net Improvement (Steering Burden)Agents12.52.2−19.4
FrontierMath Tiers 1-3 (v2) - PrivateMath93.774.4−19.3
Harvey's Legal Agent BenchmarkLegal5.423.8+18.3
Artificial Analysis Intelligence Index v5.0Indexes54.536.4−18.1
Artificial Analysis Intelligence Index v5.0.1Indexes54.636.6−18.0
Snorkel Agentic Coding 2.0Code42.825.0−17.8
Code MigrationCode65.147.4−17.7
ArXivMath 06/2026Math94.476.7−17.7
SheetBench-50Finance & Accounting60.542.8−17.7
FrontierCode 2.0 (Artificial Analysis)Code40.623.1−17.5
AA-PersonalityBench Organization indexChat & Writing81.063.8−17.3
Legal Research BenchLegal38.555.3+16.8
SnorkelLegalLegal20.736.5+15.8
CritPt v2 (agentic, avg@5)Science82.668.0−14.6
Professional Reasoning Benchmark - LegalLegal47.961.6+13.7
Professional Reasoning Benchmark - FinanceFinance & Accounting46.259.5+13.3
AA-3D v1.0 EloCode11211032−13.0
KernelBench CUDA Normalized (RTX PRO 6000)Code43.230.3−12.9
LMArena Agent Arena Praise vs ComplaintAgents27.51.7−12.9
DAYJOB: Data ScienceProfessional Work18.76.0−12.7
AA-AnalystAgent (pass^5)Professional Work50.037.5−12.5
APEX-Agents 1.1Professional Work73.261.3−11.9
NoCoT-Bench NCKI (kspine v3)General QA130113−11.7
EnterpriseBench: CoreCraftProfessional Work50.361.9+11.6
Arena Alignment IndexSafety87.977.1−10.9
DistressBenchSafety74.284.9+10.7
LiveBench 2.0Agents85.474.8−10.6
Extended NYT Connections 940 (Quadratic v1)Games95.585.1−10.4
MysteryMechanismScience46.436.0−10.4
SAFE-Teen 106 worst branch: Urgent Safety (no prompt)Safety15.425.6−10.2
Tax Agent BenchFinance & Accounting62.372.4+10.1
Westworld Finance Diligence BenchFinance & Accounting55.045.2−9.8
Medical Long Context Reasoning (MLCR-AA)Health & Medicine33.943.3+9.4
MedXpertQA MM (200-question set)Health & Medicine85.776.3−9.4
Humanity's Sixth SenseMultimodal46.637.4−9.2
τ³-Banking (Mercor run, bm25_grep retrieval)Agents43.052.2+9.2
SnorkelManufacturingProfessional Work19.610.5−9.1
DAYJOB: LawLegal31.822.8−9.0
Conceptual Reasoning IndexIndexes72.766.2−8.9
LMCAGeneral QA62.853.9−8.9
Epoch Capabilities Index (ECI)Indexes169158−8.8
AutomationBench-AA 1.0.6Agents66.657.9−8.7
SWE-bench Verified Extended (Mercor held-out)Code45.354.0+8.7
FrontierCode 1.1 Main (Artificial Analysis)Code49.140.6−8.5
Artificial Analysis Progress Index v1Indexes348324−8.4
AA-Omniscience IndexIndexes41.525.0−8.3
Humanity's Last ExamGeneral QA51.343.2−8.1
Finance Agent v2Finance & Accounting52.060.0+7.9
KernelBench Mega Normalized (RTX PRO 6000)Code14.16.2−7.9
APEX-SWECode46.939.1−7.8
Snorkel SWE-bench CLICode9.82.2−7.6
DAXBench v2Code92.685.1−7.4
DeepSWE v1.1Code75.267.8−7.4
GDP.pdfProfessional Work35.027.6−7.4
Gray Swan IPI Attack Success (k=15)Safety8.615.9−7.3
DAYJOB: AccountingFinance & Accounting25.218.0−7.2
GeneBench-Pro Public Release SubsetScience33.026.0−7.0
Legacy-Bench (Fireworks, avg@3)Code55.062.0+7.0
Access to Justice BenchmarkLegal65.872.2+6.3
Harvey LAB (Mercor run, undisclosed ~213-task cohort)Legal8.815.0+6.2
ACCoRDGeneral QA80.975.1−5.8
CritPt (avg@5)Science31.726.0−5.7
Roboflow Vision Evals (September 2026 protocol)Multimodal85.579.8−5.7
SAFE-Teen 106: Urgent Safety (no prompt)Safety3.49.0−5.6
SnorkelUnderwrite 2.0Finance & Accounting25.019.6−5.4
LMArena Agent Arena Steering BurdenAgents13.42.6−5.4
ComplexConstraints (November 2026 rescoring)Chat & Writing70.164.8−5.3
HANDBOOK.md Agents v1.1Professional Work28.633.9+5.3
SAFE-Teen 106 worst branch: Health Advice (no prompt)Safety8.813.9−5.1
ChessBench AccuracyGames80.776.0−5.0
WellnessBenchSafety66.061.0−5.0
Snorkel Digital Work IndexIndexes21.216.3−4.9
MedXpertQA MM Extended (Mercor held-out)Health & Medicine64.760.1−4.6
HiL-Bench (Human-in-Loop Benchmark)Agents62.366.9+4.6
SWE-TogetherCode58.362.4+4.1
LMArena Agent Arena Confirmed SuccessAgents11.73.7−4.0
SciCode 1.0.1 (Artificial Analysis)Code55.859.7+3.9
Ramp Accounting BenchFinance & Accounting46.442.5−3.9
Artificial Analysis Intelligence Index v4.3.2Indexes51.848.1−3.7
ChessBench CoherenceGames87.591.2+3.7
SnorkelFinance 2.0Finance & Accounting6.610.0+3.4
Terminal-Bench 2.1Code85.488.8+3.4
PCT Political Inconsistency (CAIS)Safety37.340.6−3.4
Excel Modeling BenchmarkFinance & Accounting70.867.4−3.3
CharXiv Extended (Mercor held-out)Multimodal85.181.8−3.3
GDPval-AA v2.1Professional Work16071681+3.2
DAYJOB: FinanceFinance & Accounting18.014.8−3.2
SAFE-Teen 106 worst branch: Health Advice (under-18 prompt)Safety7.74.5+3.2
Vibe Code Bench v1.1Code88.985.9−3.1
Vals Index v2.1Indexes61.258.2−3.0
Immigration BenchLegal67.865.0−2.8
LMArena Text Arena EloChat & Writing14821496+2.5
Financial Coaching BenchFinance & Accounting68.571.0+2.4
Visa Agentic Commerce BenchmarkAgents62.264.4+2.2
CharXiv (descriptive and reasoning combined)Multimodal96.093.9−2.1
SAFE-Teen 106: Health Advice (no prompt)Safety4.76.7−2.0
Harvey LAB-AA v1.1 (hallucination-gated)Legal6.98.9+1.9
LMArena Agent Arena Bash ReliabilityAgents5.02.0−1.5
AA-Briefcase v1.1 EloAgents15551581+1.5
SpeciEval (Refusal-Aware Retries, Historical)Safety97.498.8+1.4
SpeciEval (Pooled English Runs)Safety97.498.7+1.4
SAFE-Teen 106 worst branch: Urgent Safety (under-18 prompt)Safety12.711.3+1.3
SAFE-Teen 106: Health Advice (under-18 prompt)Safety3.32.1+1.3
SAFE-Teen 106 worst branch: Relationships (under-18 prompt)Safety4.53.2+1.3
SimpleBenchGeneral QA82.981.8−1.1
AA-LCR v1.1Search84.083.0−1.0
Ruby LLM Benchmarks — Program FixerCode75.074.1−0.9
OTIS Mock AIME 2024-2025Math100.099.2−0.8
CVDP (517-problem set)Code55.154.3−0.8
LMArena Agent Arena Bash RecoveryAgents4.56.1+0.8
SAFE-Teen 106: Urgent Safety (under-18 prompt)Safety2.02.7−0.7
ObviousBenchGeneral QA100.099.3−0.7
APEX-Accounting 1.1Finance & Accounting64.364.9+0.6
GDP.xlsxProfessional Work22.022.6+0.6
ProgramBenchCode3.02.5−0.5
SnorkelRevOpsProfessional Work12.112.6+0.5
SWE-bench Multimodal (Vals v1, 480 Tasks)Code43.543.1−0.4
DAYJOB: TaxFinance & Accounting4.03.6−0.4
SAFE-Teen 106: Relationships (under-18 prompt)Safety0.90.6+0.3
Vision Arena OverallMultimodal12931291−0.3
DAYJOB: HealthcareHealth & Medicine7.67.8+0.2
ForecastBench Preliminary Dataset Brier IndexForecasting61.761.9+0.2
LMArena Agent Arena Tool HallucinationAgents1.00.8−0.1
AIME 2025Math100.0100.00.0
DTBenchGeneral QA96.596.50.0

GPT-6.1 SolvsMuse Spark 1.3normalised score marginsData Nov 15, 2026Data: Vals AI, Artificial Analysis, Mercor and 41 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…