Atlas

Compare Models

Claude Opus 5.5 vs GPT-5.6 Terra

Claude Opus 5.5 leads by 7.0 capability points (≈95% interval +4.7 to +9.4): a clear lead over GPT-5.6 Terra · P(higher) ≈ 100%

Wins 150 of 192 shared benchmarks (19 lost, 23 tied)

Capability vs. relative cost
Claude Opus 5.5GPT-5.6 TerraClosed 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.

Claude Opus 5.5vsGPT-5.6 Terracapability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins150 – 19 · 192 shared, 23 tied

Comparing matching setups only (88 benchmarks excluded) · suites counted once
← Claude Opus 5.5GPT-5.6 Terra →

Claude Opus 5.5vsGPT-5.6 Terrawins by categoryData Nov 15, 2026Data: Vals AI, Mercor, Artificial Analysis and 53 othersVillage LLM Atlasatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is GPT-5.6 Terra minus Claude Opus 5.5.

VCT-Refusal (CAIS)Safety90.410.7−79.7
YC-BenchGames2873725636144−78.1
HPCT Refusal (CAIS)Safety100.025.0−75.0
LisanBench Path LengthGames240293303−64.4
LisanBench Difficulty-WeightedGames59631197−51.0
Terminal-Bench 4.0 (Artificial Analysis)Code59.610.1−49.5
Terminal-Bench-Science 0.1Science61.913.8−48.1
AutomationBench 1.0.6 (Zapier)Agents42.50.8−41.7
GameBench 2Games94.853.5−41.3
HardGeoBench Avg DistanceMultimodal6691991−40.8
LMArena Agent Arena Net Improvement (Steering Burden)Agents19.5-1.8−40.3
Concept Synth INDUCTION Challenge100 Round 1General QA65.025.0−40.0
Terminal-Bench 4.0Code64.826.3−38.6
Terminal-Bench Science 0.1 (Vals)Science48.610.0−38.6
Humanity's Last Exam RMS Calibration Error (CAIS)General QA13.850.7−36.8
APEX-SWECode67.631.1−36.5
Gray Swan IPI Attack Success (k=15)Safety1.037.3−36.3
Mystery Game PuzzlesGames71.035.0−36.0
BrowseComp Extended (Mercor held-out)Search60.525.3−35.2
Medical Long Context Reasoning (MLCR-AA)Health & Medicine66.731.7−35.0
EnigmaEvalGeneral QA57.823.1−34.7
Concept Synth INDUCTION Challenge64 Round 1General QA71.937.5−34.4
BioTIER RefusalSafety98.765.6−33.2
CadQueryEval (Corrected Geometry)Code100.068.0−32.0
MindCube TinyMultimodal93.162.6−30.5
Hedge-BenchFinance & Accounting44.213.8−30.4
Agent Red Teaming k=100 (CAIS)Safety24.454.6−30.1
Furniture Assembly BenchmarkMultimodal83.354.2−29.2
GDPval Extended (Mercor held-out rubric score)Professional Work75.847.2−28.6
SWE Atlas Codebase QnA Extended (Mercor held-out)Code43.316.0−27.3
ProofBench v1.1Math100.074.0−26.0
Humanity's Last Exam DiamondGeneral QA54.628.8−25.8
AA-Briefcase v1.1 EloAgents18011333−25.6
ChessBench EloGames1349925−25.2
HackerRank-BenchCode51.226.1−25.1
HardGeoBench <=100miMultimodal61.336.3−25.0
SpatialViz-BenchMultimodal89.765.2−24.5
FrontierMath Tier 4 (v2) - PrivateMath95.070.7−24.3
Design Arena Overall Code CategoriesCode13841238−24.2
ZeroBench (Main questions, pass@5)Multimodal43.019.0−24.0
AA-Omniscience IndexIndexes46.40.1−23.2
DAYJOB: HealthcareHealth & Medicine24.71.6−23.1
SM BenchSafety81.859.0−22.7
FrontierCode 2.0 (Artificial Analysis)Code46.423.9−22.5
HardGeoBench Country AccuracyMultimodal67.545.0−22.5
DAYJOB: FinanceFinance & Accounting23.91.8−22.1
GeneBench-Pro Public Release SubsetScience30.08.0−22.0
Humanity's Sixth SenseMultimodal44.622.7−21.9
APEX-AccountingFinance & Accounting61.840.3−21.5
LMArena Agent Arena Praise vs ComplaintAgents40.3-1.9−21.1
ALE-BenchCode27482154−21.0
ZeroBench Main Mean pass@1 (five samples)Multimodal35.014.2−20.8
SWE-bench VerifiedCode98.277.8−20.4
Blueprint-Bench 2Multimodal0.50.3−20.4
MedXpertQA MM Extended (Mercor held-out)Health & Medicine71.251.0−20.2
Chess Puzzles (task version 2.0)Games67.047.0−20.0
Harvey's Legal Agent Benchmark - Energy Natural ResourcesLegal20.00.0−20.0
Harvey's Legal Agent Benchmark - Healthcare Life SciencesLegal20.00.0−20.0
Harvey's Legal Agent Benchmark - Structured Finance SecuritizationLegal0.020.0+20.0
LAB-Bench LitQA2 (SecureBio)Science71.151.1−20.0
BioSecBench-RespiratoryVirus-Surveillance V1 (110 evaluations)Science25.45.8−19.7
ProgramBench (Mercor 100-task subset)Code20.01.0−19.0
VCT v2 (SecureBio)Science60.541.9−18.6
BenchCAD - geometric overlapMultimodal96.278.2−18.0
GDPval-AA v2.1Professional Work18661461−17.9
HardGeoBench <=10miMultimodal42.525.0−17.5
PCT Political Inconsistency (CAIS)Safety29.146.2−17.1
LAB-Bench ProtocolQA (SecureBio)Science94.077.0−17.0
HieroglyphBenchMultimodal37.420.5−16.9
TextQuests Game ProgressGames57.340.5−16.8
Conceptual Reasoning IndexIndexes77.965.7−16.8
CursorBench 4.0Code57.841.3−16.5
τ³-Banking (Mercor run, bm25_grep retrieval)Agents55.038.8−16.2
Roboflow Vision Evals (October 2026, five tasks)Multimodal82.066.0−16.0
Humanity's Last Exam (Text-Only)General QA57.541.9−15.6
ReactBench (39 tasks)Code54.439.0−15.4
ZeroBench Main Questions pass^5Multimodal25.010.0−15.0
Agents' Last Exam Overall ScoreAgents63.248.5−14.7
SUPERChemScience79.664.9−14.7
VCT (SecureBio)Science61.447.2−14.2
LMArena Agent Arena Confirmed SuccessAgents20.7-7.2−13.9
ACCoRDGeneral QA88.474.7−13.8
FACTS Benchmark SuiteGeneral QA65.352.0−13.3
LMCAGeneral QA68.255.0−13.3
CVDP (517-problem set)Code64.150.9−13.2
ITBench-AAAgents38.251.0+12.8
SWE-bench Verified Extended (Mercor held-out)Code66.078.7+12.7
CritPt v2 (agentic, avg@5)Science81.769.7−12.0
SciCode 1.0.1 (Artificial Analysis)Code66.955.0−11.9
Senior SWE-BenchCode39.327.4−11.9
MBCT (SecureBio)Science68.756.9−11.8
Roboflow Vision Evals (September 2026 protocol)Multimodal85.573.8−11.7
LLM Chess EloGames15751293−11.6
Bio Capabilities Index (SecureBio)Indexes171155−11.6
LiveBench 2.0Agents86.575.0−11.5
ERQAMultimodal75.163.6−11.5
SciCode (main problems)Science27.315.8−11.5
Terminal-Bench 2.1 Extended (Mercor held-out)Code47.536.4−11.1
SpeciEval (Refusal-Aware Retries, Historical)Safety87.298.1+10.8
SpeciEval (Pooled English Runs)Safety87.298.1+10.8
GBENCH One-shot Coding GScoreGames73.963.3−10.6
ChessBench CoherenceGames80.169.7−10.4
Extended NYT Connections 940 (Quadratic v1)Games88.578.4−10.1
MedXpertQA MM (200-question set)Health & Medicine85.775.7−10.0
Creative Writing v3 EloChat & Writing20511856−9.8
Excel Modeling BenchmarkFinance & Accounting75.966.2−9.7
LMArena Agent Arena Steering BurdenAgents18.6-0.5−9.6
APEX-Agents 1.1Professional Work81.371.8−9.5
ARC-AGI-2General QA93.383.9−9.4
DeepSWE v1.1Code74.264.8−9.4
BullshitBench v2 LeaderboardSafety63.054.0−9.0
Machiavelli Deception (CAIS)Safety77.580.4−8.7
MedScribeHealth & Medicine91.482.9−8.6
Harvey LAB (Mercor run, undisclosed ~213-task cohort)Legal10.82.3−8.5
ChessBench AccuracyGames79.371.9−7.9
BuseyBench SVGMultimodal6.86.1−7.7
GDPval (Mercor 213-task rubric score)Professional Work88.380.7−7.6
DeepsecBench — Revalidated V1Code26.719.1−7.5
DeepsecBench — Revalidated V1 (233 issues)Code26.619.1−7.5
PubMedQA (SecureBio)Health & Medicine87.980.4−7.5
LMArena Agent Arena Bash ReliabilityAgents15.10.7−7.2
MMMU ProMultimodal86.679.5−7.1
LMArena Agent Arena Bash RecoveryAgents15.00.9−7.1
IntPhys 2Multimodal68.361.3−7.0
LMArena Text Arena EloChat & Writing15061468−6.7
TextQuests Average Cumulative HarmSafety24.222.2+6.7
Longform Creative WritingChat & Writing84.678.0−6.6
WMDP-Bio (SecureBio)Science96.189.7−6.4
MedCodeHealth & Medicine49.843.4−6.4
Epoch Capabilities Index (ECI)Indexes169161−6.3
DeepResearch Bench II (Mercor 130-task set)Search53.347.2−6.1
RedlineBenchLegal42.047.8+5.8
AA-LCR v1.1Search84.779.0−5.7
DTBenchGeneral QA98.993.3−5.6
GPQA-Bio (SecureBio)Science90.985.3−5.6
Olam Social Poker EloGames15061511+5.4
MMLU ProGeneral QA93.488.1−5.3
WCB (SecureBio)Science54.559.8+5.3
FrontierMath Tiers 1-3 (v2) - PrivateMath91.286.0−5.3
Financial Coaching BenchFinance & Accounting76.271.0−5.2
AdvancedIFChat & Writing82.878.1−4.7
Terminal-Bench 2.1Code83.588.0+4.5
SAGE v1.1Math91.387.0−4.3
GDP.pdf (Artificial Analysis, All-pass)Professional Work28.824.6−4.2
Harvey HLAB — Criteria Pass RateLegal75.879.8+4.0
MASKSafety89.493.3+3.9
SAB ResearchFinance & Accounting97.293.4−3.9
APEX-1 (Mercor, mean score)Professional Work65.869.5+3.7
SWE Atlas - Codebase QnACode48.945.2−3.7
CharXiv Extended (Mercor held-out)Multimodal84.781.1−3.6
MMMU-Pro Extended (Mercor held-out)Multimodal51.047.4−3.6
Ruby LLM Benchmarks — Program FixerCode77.574.0−3.6
Quant-Arena (iterative total score)Finance & Accounting44.347.6+3.3
RiemannBenchMath69.666.4−3.2
SAB RankingForecasting-0.1-0.0+3.0
AHK-EvalCode91.788.9−2.8
BrowseComp (Mercor 130-question subset)Search88.585.8−2.7
CharXiv (descriptive and reasoning combined)Multimodal96.494.1−2.3
ObviousBenchGeneral QA100.097.9−2.1
ARC-AGI-1 Semi-PrivateGeneral QA98.596.5−2.0
MMLU-Bio (SecureBio)Science92.790.7−2.0
CritPt (avg@5)Science31.730.0−1.7
SAGEMath45.847.0+1.2
SWE-bench MultilingualCode95.594.4−1.1
DistressBenchSafety75.676.4+0.9
Creative Writing v3 Rubric ScoreChat & Writing16.015.9−0.4
OTIS Mock AIME 2024-2025Math100.099.7−0.3
AA-LCR Extended (Mercor held-out)Search74.274.4+0.2
LMArena Agent Arena Tool HallucinationAgents0.71.0+0.1
AIME 2025Math100.0100.00.0
Eval Connections One-Shot with TrapsGames86.086.00.0
Harvey's Legal Agent Benchmark - Antitrust CompetitionLegal0.00.00.0
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 - Capital MarketsLegal0.00.00.0
Harvey's Legal Agent Benchmark - Corporate GovernanceLegal0.00.00.0
Harvey's Legal Agent Benchmark - Corporate M&ALegal0.00.00.0
Harvey's Legal Agent Benchmark - Data Privacy CybersecurityLegal0.00.00.0
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 - Environmental ESGLegal0.00.00.0
Harvey's Legal Agent Benchmark - Funds Asset ManagementLegal0.00.00.0
Harvey's Legal Agent Benchmark - ImmigrationLegal0.00.00.0
Harvey's Legal Agent Benchmark - InsuranceLegal0.00.00.0
Harvey's Legal Agent Benchmark - Intellectual PropertyLegal0.00.00.0
Harvey's Legal Agent Benchmark - International Trade SanctionsLegal0.00.00.0
Harvey's Legal Agent Benchmark - Litigation Dispute ResolutionLegal0.00.00.0
Harvey's Legal Agent Benchmark - Real EstateLegal0.00.00.0
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

Claude Opus 5.5vsGPT-5.6 Terranormalised score marginsData Nov 15, 2026Data: Vals AI, Mercor, Artificial Analysis and 53 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…