Atlas

Compare Models

GPT-5.5 vs Grok 4.7

Grok 4.7 leads by 0.6 capability points (≈95% interval −1.5 to +2.6): statistically tied · P(higher) ≈ 71%

GPT-5.5 wins 86 of 169 shared benchmarks (62 lost, 21 tied)

Capability vs. relative cost
GPT-5.5Grok 4.7Closed 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-5.5vsGrok 4.7capability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins86 – 62 · 169 shared, 21 tied

Comparing matching setups only (63 benchmarks excluded) · suites counted once
← GPT-5.5Grok 4.7 →

GPT-5.5vsGrok 4.7wins by categoryData Nov 15, 2026Data: Vals AI, Mercor, Artificial Analysis and 40 othersVillage LLM Atlasatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is Grok 4.7 minus GPT-5.5.

FrontierMath Tier 4 (v2) - PrivateMath72.517.1−55.4
Harvey's Legal Agent Benchmark - Capital MarketsLegal0.040.0+40.0
LAB-Bench LitQA2 (SecureBio)Science49.816.9−32.9
FrontierMath Tiers 1-3 (v2) - PrivateMath85.353.0−32.3
LLM Chess EloGames1532822−29.1
AA-Briefcase v1.1 EloAgents11351636+27.4
Mystery Game PuzzlesGames56.029.0−27.0
LAB-Bench ProtocolQA (SecureBio)Science77.951.4−26.5
Terminal-Bench 2.1 Extended (Mercor held-out)Code38.011.8−26.2
FoodTruck BenchGames6140842561−24.2
ARC-AGI-2General QA85.061.4−23.6
Furniture Assembly BenchmarkMultimodal44.220.8−23.3
LLM Creative Story WritingChat & Writing2.40.4−22.6
Chess Puzzles (task version 2.0)Games63.041.0−22.0
SM BenchSafety65.486.3+20.9
BrowseComp Extended (Mercor held-out)Search19.039.5+20.5
Harvey's Legal Agent Benchmark - Antitrust CompetitionLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Corporate M&ALegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Data Privacy CybersecurityLegal20.00.0−20.0
Harvey's Legal Agent Benchmark - Funds Asset ManagementLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Healthcare Life SciencesLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Intellectual PropertyLegal0.020.0+20.0
Harvey's Legal Agent Benchmark - Litigation Dispute ResolutionLegal0.020.0+20.0
Extended NYT Connections 940 (Quadratic v1)Games96.276.8−19.4
AutomationBench-AA 1.0.6Agents47.365.6+18.3
Vending-Bench 2Games752410537+18.2
LAB-Bench CloningScenarios (SecureBio)Science97.380.1−17.2
EnterpriseBench: CoreCraftProfessional Work51.368.2+16.9
APEX-SWECode36.653.4+16.8
ArXivMath 05/2026Math77.560.8−16.7
BrokenArXiv 06/2026Math69.452.8−16.7
Vibe Code Bench v1.1Code69.886.2+16.3
ChartographyMultimodal31.014.7−16.3
ChessBench EloGames13491076−16.2
YC-BenchGames12060451666389+16.1
SWE-bench Verified Extended (Mercor held-out)Code39.355.3+16.0
TaxCalcBench TY25Finance & Accounting24.08.0−16.0
GDPval-AA v2.1Professional Work13571717+15.9
BuseyBench SVGMultimodal6.04.6−14.3
Olam Social Poker EloGames15211508−14.1
ChessBench CoherenceGames100.086.3−13.7
ArXivMath 06/2026Math83.670.1−13.5
LLM Debate BenchmarkChat & Writing15281620+13.1
HPCT (SecureBio)Science66.853.7−13.1
Toloka Arena (v1)Agents52.339.3−13.0
GameBench 2Games59.972.3+12.4
Bio Capabilities Index (SecureBio)Indexes158141−12.1
CadQueryEval (Corrected Geometry)Code76.088.0+12.0
Terminal-Bench 4.0 (Artificial Analysis)Code14.625.8+11.1
MBCT (SecureBio)Science57.146.1−11.0
SAGEMath51.540.8−10.7
Harvey HLAB — Criteria Pass RateLegal80.090.5+10.5
Code2Video-AA EloCode10001127+10.3
DeepsecBench — Revalidated V1Code21.110.9−10.2
DeepsecBench — Revalidated V1 (233 issues)Code21.010.9−10.1
ALE-BenchCode21081823−10.1
BrokenArXiv 05/2026Math50.040.0−10.0
Ruby LLM Benchmarks — Program FixerCode77.668.0−9.6
Vision Arena OverallMultimodal12831249−8.8
MedXpertQA MM Extended (Mercor held-out)Health & Medicine60.852.0−8.8
Tax Agent BenchFinance & Accounting60.569.0+8.5
Creative Writing v3 EloChat & Writing18452012+8.5
Artificial Analysis Intelligence Index v4.3.2Indexes38.446.4+8.1
MedXpertQA MM (200-question set)Health & Medicine76.769.0−7.7
CritPt (avg@5)Science25.418.0−7.4
SimpleQA Verified (Epoch anti-abstention)General QA63.056.0−7.0
LMArena Text Arena EloChat & Writing14811442−6.9
Legal Research BenchLegal40.447.1+6.7
DeepSWE v1.1Code64.471.0+6.6
SWE-bench MultilingualCode77.784.2+6.5
ACCoRDGeneral QA80.574.0−6.5
GBENCH One-shot Coding GScoreGames62.256.3−5.9
Conceptual Reasoning IndexIndexes67.363.1−5.8
KernelBench Mega Normalized (RTX PRO 6000)Code12.218.0+5.8
AA-Omniscience IndexIndexes20.532.0+5.8
Finance Agent v2 - Disclosure AnalysisFinance & Accounting58.964.5+5.6
RedlineBenchLegal44.149.5+5.4
LiveBench 2.0Agents68.774.1+5.4
Finance Agent v2 - Financial ModelingFinance & Accounting22.917.7−5.3
Finance Agent v2 - General Quantitative AnalysisFinance & Accounting73.768.6−5.2
ChessBench AccuracyGames78.874.0−5.1
GPQA-Bio (SecureBio)Science87.482.5−4.9
LMCAGeneral QA54.349.4−4.9
ARC-AGI-1 Semi-PrivateGeneral QA95.090.2−4.8
Longform Creative WritingChat & Writing78.283.0+4.8
Public Benefits Bench v1.1Professional Work60.965.6+4.7
CharXiv Extended (Mercor held-out)Multimodal82.778.2−4.5
AA-LCR v1.1Search84.380.0−4.3
Humanity's Last Exam DiamondGeneral QA27.923.6−4.3
DAXBench v2Code88.392.6+4.3
Finance Agent v2 - PrecedentsFinance & Accounting21.925.9+4.1
OncologyBench Strict (25 tasks, October 9 rubric)Health & Medicine14.718.7+4.0
ITBench-AAAgents45.842.1−3.7
Blueprint-Bench 2Multimodal0.40.3−3.7
DeepResearch Bench II (Mercor 130-task set)Search45.449.0+3.6
VCT v2 (SecureBio)Science45.342.0−3.3
Epoch Capabilities Index (ECI)Indexes160156−3.2
LMArena Agent Arena SteerabilityAgents6.90.5−3.2
GDP.pdfProfessional Work26.022.8−3.2
Finance Agent v2 - General Qualitative AnalysisFinance & Accounting65.668.7+3.2
MMMU ProMultimodal81.278.2−3.0
APEX-1 (Mercor, mean score)Professional Work68.465.4−3.0
Roboflow Vision Evals (September 2026 protocol)Multimodal74.871.9−2.9
Vals Index v2Indexes57.460.2+2.8
Medical Long Context Reasoning (MLCR-AA)Health & Medicine17.815.0−2.8
ForecastBench Preliminary Dataset Brier IndexForecasting63.360.7−2.6
CharXiv (descriptive and reasoning combined)Multimodal94.692.0−2.6
Finance Agent v2 - All PassFinance & Accounting39.642.1+2.6
MedScribeHealth & Medicine86.989.4+2.5
AA-AnalystAgent (pass^5)Professional Work50.052.5+2.5
Roboflow Vision Evals (October 2026, five tasks)Multimodal66.063.5−2.5
Excel Modeling BenchmarkFinance & Accounting64.567.0+2.4
AA-LCR Extended (Mercor held-out)Search70.568.2−2.3
Design Arena Overall Code CategoriesCode12571270+2.2
LegalBenchLegal86.584.4−2.1
Senior SWE-BenchCode29.527.4−2.1
WCB (SecureBio)Science59.757.6−2.1
GPQA DiamondGeneral QA90.792.7+2.1
SWE Atlas Codebase QnA Extended (Mercor held-out)Code17.315.3−2.0
GeneBench-Pro Public Release SubsetScience8.010.0+2.0
OTIS Mock AIME 2024-2025Math100.098.1−1.9
BrowseComp (Mercor 130-question subset)Search77.775.8−1.9
Humanity's Last Exam (Text-Only)General QA45.043.1−1.9
SpeciEval (Pooled English Runs)Safety98.296.4−1.8
SpeciEval (Refusal-Aware Retries, Historical)Safety98.296.4−1.8
SWE Atlas - Codebase QnACode45.247.0+1.8
SciCode 1.0.1 (Artificial Analysis)Code56.157.8+1.6
APEX-Agents 1.1Professional Work70.368.7−1.6
Finance Agent v2 - Market AnalysisFinance & Accounting60.361.8+1.5
APEX-AccountingFinance & Accounting48.246.7−1.5
Finance Agent v2 - ComparablesFinance & Accounting43.742.2−1.4
Finance Agent v2 - Earnings AnalysisFinance & Accounting73.274.6+1.4
ARC-AGI-3 (Semi-Private)General QA0.41.7+1.3
PubMedQA (SecureBio)Health & Medicine80.579.2−1.3
LMArena Agent Arena Bash RecoveryAgents9.57.2−1.2
AdvancedIFChat & Writing81.380.2−1.1
Artificial Analysis Progress Index v1Indexes320323+0.9
LMArena Agent Arena Net ImprovementAgents4.24.0−0.7
Terminal-Bench 2.1Code84.385.0+0.7
ObviousBenchGeneral QA100.099.3−0.7
ComplexConstraintsChat & Writing49.548.9−0.6
Finance Agent v2 - AdjustmentsFinance & Accounting45.746.3+0.5
MedCodeHealth & Medicine49.149.6+0.5
Code MigrationCode45.244.8−0.3
Creative Writing v3 Rubric ScoreChat & Writing15.915.9+0.3
SWE-bench VerifiedCode78.879.1+0.3
MineBench — Bradley-Terry RatingGames18461848+0.1
MMLU ProGeneral QA88.988.8−0.1
AIME 2025Math100.0100.00.0
DTBenchGeneral QA96.096.00.0
FrontierMath ErdősMath0.00.00.0
Harvey's Legal Agent Benchmark - Arbitration International Dispute ResolutionLegal0.00.00.0
Harvey's Legal Agent Benchmark - Banking FinanceLegal0.00.00.0
Harvey's Legal Agent Benchmark - Bankruptcy RestructuringLegal0.00.00.0
Harvey's Legal Agent Benchmark - Corporate GovernanceLegal0.00.00.0
Harvey's Legal Agent Benchmark - Emerging Companies Venture CapitalLegal0.00.00.0
Harvey's Legal Agent Benchmark - Employment LaborLegal0.00.00.0
Harvey's Legal Agent Benchmark - Energy Natural ResourcesLegal20.020.00.0
Harvey's Legal Agent Benchmark - Environmental ESGLegal0.00.00.0
Harvey's Legal Agent Benchmark - ImmigrationLegal0.00.00.0
Harvey's Legal Agent Benchmark - InsuranceLegal0.00.00.0
Harvey's Legal Agent Benchmark - International Trade SanctionsLegal0.00.00.0
Harvey's Legal Agent Benchmark - Real EstateLegal0.00.00.0
Harvey's Legal Agent Benchmark - Structured Finance SecuritizationLegal0.00.00.0
Harvey's Legal Agent Benchmark - TaxLegal0.00.00.0
Harvey's Legal Agent Benchmark - Trusts Estates Private ClientLegal0.00.00.0
Harvey's Legal Agent Benchmark - White Collar Defense InvestigationsLegal0.00.00.0
ProgramBenchCode0.50.50.0
SRE Bench — Fully SolvedCode3.83.80.0

GPT-5.5vsGrok 4.7normalised score marginsData Nov 15, 2026Data: Vals AI, Mercor, Artificial Analysis and 40 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…