Benchmark directory
Every benchmark in the catalog, each with its own page covering what it tests, who published it, and which scores are on record. The interactive dashboard is here.
3228 benchmarks
- $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
- 360CityArena
- 3DCodeBench
- 4DSynth-Nav
- 5-Dialects-BN
- A Large-Scale Dataset and Benchmark
- A-Bench
- A-OKVQA
- A3S-Bench
- AA-AnalystAgent
- AA-Briefcase
- AA-Briefcase
- AA-Index
- AA-LCR
- AA-LCR
- AA-LCR
- AA-Omniscience Accuracy
- AA-Omniscience Index
- AA-Omniscience Non-Hallucination Rate
- AARRI-Bench
- ABC-Bench
- AbductionRules
- AbsPyramid
- ACEBench
- ActBench
- ActionAtlas
- Active-SWE
- ActiveFly-Bench
- ActiveSciBench
- ActiveVision
- ActivityNet
- AD-Diff Bench
- Ada-LEval
- AdaMem-Bench
- AdaPlanBench
- Adaptive Adversaries
- ADeptS-Bench
- AdvancedIF
- AdvancedMathBench
- AdversaBench
- AdvPlan-Bench
- AE-UAV
- AECBench
- AEGBench
- ael_gate_benchmark_cases_template
- AeroCopilotBench
- AeroGround
- AetherCode
- AFDBench
- AFFORDANCE20Q
- AFQMC
- AFT-Bench
- AGC-Bench
- Agent App Benchmark
- Agent Memory Benchmark
- Agent Planning Benchmark
- Agent Retrieval Bench
- Agent Startup Bench
- Agent Systems Benchmark (ASB)
- agent-injection-bench
- agent-model-bench
- AgentAbstain
- AgentBoard
- AGENTCHAOSBENCH
- AgentCIBench
- AgentDrift
- AgentFairBench
- AgentHarm
- AgentHazard
- AgentHijack
- AgentHPOBench
- Agentic Evaluation in 2D Mazes
- Agentic Infrastructure ChangeBench
- Agentic NetDevOps Benchmark
- Agentic Security Benchmark
- AgenticDataBench
- AgenticInterpBench
- AgenticSTS
- AgentInfraBench
- AgentJudgeBench
- AgentLens
- AgentMemBench
- AgentRedBench
- AgentRelBench
- AgentRewardBench
- Agents' Last Exam
- Agents' Last Exam
- Agents' Last Exam (ALE)
- AgentS4D
- AgentSLABench
- AgentSysBench
- AgentWorldBench
- AGI Maze Prediction Datasets and Benchmark
- AGIEval
- AGIEval
- AGIEval
- AgingBench
- AgMMU
- AGORA
- AGOS-Bench
- AgroTools
- AgroVG
- AGVBench
- AI Coding Agent Water Simulation Benchmark
- AI gateway reproducible benchmark
- AI Security Leaderboard
- AI2 Reasoning Challenge (ARC)
- AI2D
- AI4AI-Bench
- AIB
- AICompanionBench
- AIDABench
- Aider
- Aider Polyglot
- Aider-Polyglot
- Aider-Polyglot Edit
- AIME
- AIME
- AIME 2024
- AIME 2025
- AIME 2025
- AIME 2026
- AIR-BENCH
- AIR-Bench
- AIR-Bench
- AIRBench-2024
- AirGroundBench
- AIRTBench
- AISE-Bench
- AITZ_EM
- ALD/E-ImageMiner
- ALE-Bench
- alem
- AlgoWorlds
- AlignBench
- AlignBench
- Alipay-PIBench
- AllocBench
- Almieyar-Oryx-BloomBench
- AlpacaEval 2.0
- alzheimer-classification-benchmark
- AM-Bench
- AMBROSIA
- AMC_2022_23
- AmchiBias
- AMNESIA
- AMO Bench
- amo-bench-p
- AMPBench-MT
- AMSbench
- AnchorBench
- Ancient-Bench
- Android Control High_EM
- Android Control Low_EM
- AndroidBench
- AndroidDaily
- AndroidWorld
- AndroidWorld_SR
- Aneumo
- Animation2Code
- AnnoBench
- AnnoMI Counselling Dialogue Analysis
- AntiSkillBench
- Any2Poster Bench
- AnyGroundBench
- AOR-Bench
- Apex
- APEX-Accounting
- APEX-Agents
- APEX-Agents
- APEX-Agents-AA
- APEX-SWE
- API-Bank
- APIFlow-Bench
- apitest
- APPS
- AppSim-Bench
- APS-Bench
- AQUA-RAT
- ARAC-Bench
- ArbiGraph
- Arc
- ARC-AGI
- ARC-AGI
- ARC-AGI v2
- ARC-AGI-2
- ARC-AGI-3
- ARC-AGI-3
- ARC-C
- ARC-c
- ARC-E
- ARC-e
- ArcAGI2
- ArchEGraph
- Arena Hard
- Arena-Hard
- Arena-Hard v2
- Arena-Hard-Auto
- ArgusInspection
- ARKitScenes
- ArmnetBench
- ART-SafeBench
- ArtECulture
- Artifacts Bench
- ArtifactsBench
- Artificial Analysis
- ArXivMath
- ArXivMath
- ASDiv
- ASI-Bench
- ASI-Bench
- ASOB-Bench
- ASPIRE
- ASR Model Benchmark
- AssayBench-Loop
- AssertLLM2
- AssetOpsBench
- AstroMind
- Asuka-Bench
- AsyncTool
- AtelierEval
- atlas
- AtmosCoder-Bench
- ATOM-Bench
- Atrex-Bench
- AttaQ
- AuAu
- AudioJailbreak
- AudioTrust
- AuthBench
- AuthMem-Bench
- AuthorityBench
- AuthTrace
- AutoAdvExBench
- AutoLab
- AutoLogi
- AutomationBench
- AutomationBench
- AutomationBench-AA
- AutomationBench-AA
- AutoMedBench
- AutoWorldModel-Bench
- AV-Odyssey-Bench
- AV-SafetyBench
- AV-SyncBench
- AvalancheBench
- Avalon-ToM-Bench
- AVBench
- AVCap-Bench
- AVE-Compass
- AVI-Bench
- AVShift
- AX-b
- AX-g
- AXBENCH
- AXIS
- BABE
- BabelJudge
- BABILong
- BabyVision
- BabyVision
- BackendForge
- Backtrader-Bench
- BAITBENCH
- BALMS
- BanglaSafe
- BanglaVeilGuard
- BanglaWild
- BankerToolBench
- BankerToolBench
- BasketballBench
- BavGround
- BBH
- BBH
- BC-VL
- Beam 128K
- BEAR-Bench
- Behavior-Skill
- Behavior2Trip
- BehaviorBench
- BehaviorBench
- BekchiAI-Benchmark
- Belebele
- BELLS-O
- BenchBench-Protocol
- BenchCAD
- BenchCAD (with Python tool)
- BenchMAX
- BenSyc
- Beyond AIME
- Beyond Fluency: A Rubric-Based Benchmark for Evaluating Saudi Dialect and Cultural Competence in Large Language Models
- BeyondAIME
- BeyondMasks
- BFCL
- BFCL
- BFCL v2
- BFCL-v3
- BFCL-V4
- BFCL_v3_MultiTurn
- BFI-Adapt
- BG-REAL
- Big Bench Audio
- Big Finance Bench
- BIG-Bench
- BIG-Bench Extra Hard
- BIG-Bench Extra Hard
- BIG-Bench Hard
- BigCodeBench
- BigCodeBench
- BigCodeBench-Full
- BigCodeBench-Hard
- BigFinanceBench
- BigOBench
- BilliardPhys-Bench
- BIM-Edit
- BinJudgeBench
- BioLP-Bench
- BioMysteryBench
- BioMysteryBench
- BioSecBench-Refusal
- BioSecBench-Surveillance
- BIRD-History
- Bird-SQL (dev)
- BixBench
- BixBench3
- Blind-Spots-Bench
- BLINK
- BLINK
- BlueFin
- Blueprint-Bench 2
- Blueprint-Bench 2
- BLUEX v2
- Boiling the Frog
- BoolQ
- BoolQ
- Boundary-Bench
- BrailleBench
- BrainBench
- BRIGHT
- BrokenArXiv
- BrowseComp
- BrowseComp
- BrowseComp Long Context 128k
- BrowseComp Long Context 256k
- BrowseComp-VL
- BrowseComp-ZH
- BrowseComp-zh
- BTS-AgentBench
- BuddyVQA
- BugSourceBench
- BulkPR-Bench
- BullingerDB
- BUST
- ByteMorph
- C-Eval
- C-Eval
- C-FAITH
- C-SafeQA
- C-SUITEBENCH
- C3
- C3-Bench
- CADEngBench
- CADP-Bench
- CAFOSat
- CalArena
- CallBench
- CaLM
- CalorieBench-80K
- CamChoreo
- Candor-LR
- CanLegalRAGBench
- CANN Bench
- CAP
- CAP-Correctness
- CapProbe
- CapQuiz
- CapRiCorn-1K
- Capture-the-Flag Challenges (Internal)
- CaptureGuide-Bench
- CardioLens
- CARE-Bench
- CAREBench
- CaST-Bench
- CASTER-Bench
- CatchBench
- Causal-Plan-Bench
- CausalDS
- CausalGame
- CausalVQA
- CBNSL
- CBX-Bench
- CC-Bench-V2 Backend
- CC-Bench-V2 Frontend
- CC-Bench-V2 Repo Exploration
- CC-OCR
- CDAF — Cached Descriptive Asset Files
- CDR-Bench
- CEB
- CEO-Bench
- CEO-Bench
- CFAgentBench
- CFEval
- CFinBench
- CFLUE
- CFM-Bench
- CG-Bench
- ChainSWE
- ChaosBench-Logic
- CharadesSTA
- CHARM
- ChartAct
- ChartAnno
- ChartArena
- ChartMuseum
- Chartography
- Chartography
- ChartQA
- ChartQA
- ChartQAPro
- ChartSync
- CharXiv Reasoning
- CharXiv-D
- CharXiv-R
- CHASE-Code
- Chatbot Arena
- Chehre
- Chem World
- Chem2Gen-Bench
- ChemBench
- ChemCoTBench-V2
- ChemE ML Benchmarks
- Chess-World-Model
- CheXpert CXR
- CHIARO
- CHID
- ChildEval
- ChildGaze
- ChitraMiti
- ChLogic
- ChoroplethMap-Bench
- ChronoBench
- ChronoMagic-Bench
- ChronoPhyBench
- ChronoQG
- ChronoState
- ChurnBench
- CI Memories Coverage
- CI Memories Violation Rate
- CIAware-Bench
- CircuitReason-1k
- CITBench
- City3D-MultiGen
- CITYREP
- CityTrajBench
- CivBench
- CL-bench
- CL-bench (Life)
- ClarifyCodeBench
- ClassicLogic
- CLAW-4L
- Claw-Anything
- Claw-Eval
- Claw-SWE-Bench
- ClawArena-Team
- ClawEval-MM
- ClawProBench
- ClawTrack
- ClawTrojan
- CLBench-V
- CleanPatrick
- clembench
- CLEVER
- ClimateViz
- ClinEnv
- ClinHallu
- ClinicalMC
- CliniCARE-Bench
- ClinLens
- ClinOCR-Bench
- CLIP-CC-Bench
- CLIR-Bench
- CloningScenarios
- CLOSER-Bench
- Clouda OCR Benchmark
- CloudCons
- CLUBench
- ClueAegis-Bench
- CLUEWSC
- CMB
- CMMLU
- CMMLU
- CMNIE
- CMNLI
- CMT-Benchmark
- CN-NewsTTS Bench
- CNeo-Bench
- CNMO 2024
- CoCoBench
- COCOTree
- COD10K-C
- CODA-BENCH
- Code-QA-Bench
- CodeAssay
- CodeCriticBench
- CodeElo
- Codeforces
- CodeForces
- CodegenBench
- CodeGolf Bench
- Codegolf v2.2
- CodeMMLU
- CodeRouterBench
- CoffeeBench
- CogCanvas
- CogManip
- Cohere Agentic Question Answering
- Cohere Data Analysis
- Cohere Memory Usage Quality
- CoinVE-Bench
- CollabBench
- COLLIE
- COLLIE
- Collie
- ColorBench
- Colosseum V2
- ComBench
- CombEval
- CombiBench
- ComboShoppingBench
- CoMedBench
- CoMET-Bench
- COMEX
- Common Voice 15
- CommonSenseQA
- CommonSenseQA
- CommunityFact
- CompanionBench
- Company Mental Model Benchmark
- CompBench
- COMPINT
- ComplexConstraints
- ComplexFuncBench
- ComplexityWorld
- CompSkillBench
- ConceptEdit-Bench
- CondVLN
- ConfBench
- ConfidenceBench
- ConflictBench
- ConlangBench
- ConnectomeBench2
- Connectors
- ContactWorld
- context-flip evaluation
- ContextBench
- ContextEcho
- ContextShift
- ContextualJudgeBench
- ContextWeave
- Continual Learning Bench
- ContinuityBench
- ContinuousBench
- ContPhy
- ContractScrub
- ConVBench
- ConvBench
- COPA
- Copper Tube Defect Dataset
- Copyright-Bench
- CopyShield
- CORAL
- CORAL
- CordisBench
- CorporateBench
- CorpusQA
- CorpusQA 1M
- CountBench
- Counter-GEO-Bench
- Counterfactual-World (CF-World)
- CountQA
- CoVEBench
- CoVoST2
- CoVoST2 en-zh
- CoWorkBench
- CPI-Bench
- CPRet
- CRA-Bench
- CRAB-Bench
- CrackedPDFs
- CraftBench
- CRAG
- CRAG
- CRAG-MM-Diagnostics
- Creation-MMBench
- Creative Writing v3
- CreativeWork
- CREW-WILDFIRE
- CriticalSCM-Bench
- CriticBench
- CritPt
- CritPt
- CritPT
- CRONOS
- Cross-Engine Humanoid Benchmark
- CrossVid
- CrossView
- CrossWordBench
- Crows-Pairs
- CRPE
- CRPErelation
- CRS-Bench
- CruiseBench
- CRUX-O
- CRUXEval-Input-CoT
- CruxEval-O
- CRUXEval-Output-CoT
- CrystalXRD-Bench
- CS-Bench
- CS-Eval
- CSimpleQA
- CSL
- CSTS
- CSTutorBench
- CSVFidelity-Bench
- CTBench
- CTIBench
- CUA-HandCrafted
- CulMind
- Cultural Moment Benchmark
- CulturalMenuBench
- CULTURE-MT
- CultureForest
- CultureTalk-ID
- CultureVidBench
- CursorBench
- CursorBench v3.2
- CutCraft
- CV-Arena
- CV-FSS
- CVDP
- CVE-Bench
- CVQA
- CVSBench
- CVTG-2K
- CXR-Retrieve
- CyBench
- CyberChainBench
- CyberGym
- CyberGym
- CyberGym-E2E
- CyberMaskQA
- CyberSecEval
- CyberSecEval 4
- Cybersecurity CTFs
- D2VBench
- D3-Omni
- DABstep
- DailyBench
- DailyOmni
- DailyReport
- DAS-Bench
- DasanCallDial
- DataClawEval
- DataComp for VLMs (DCVLM)
- DataGovBench
- DataKernelBench
- DataSpace
- DBA-Bench
- DBLifeBench
- DDBench
- DDX-TRACE
- DebtBench
- DebugBench
- DECK-Bench
- DECODEM
- DeCompBench
- Deep4ge
- DeepBias
- DEEPCHART
- Deepfake-Eval-2024
- DeepPlanning
- DeepResearchBench
- DeepSearchQA
- DeepSearchQA
- DeepSWE
- DeepSWE
- DeepSWE
- DeepSWE 1.0
- DeepSWE 1.1
- DeFAb
- DeFiFlowBench
- DelegateCI-Bench
- DelistBench
- DeltaML-Bench
- DelveRL
- DEMM-Bench
- dense500
- DEPBENCH
- DeployBench
- DermMCQA
- Design2Code
- DeskCraft
- Desktop-Delta Bench
- DevDataBench
- DevicesWorld
- DeXposure-Bench
- DexVerse
- DFAH-Bench
- dgx-spark-llm-lab
- DiagChain
- Diagram-MMU
- DICE-BENCH
- DiCoBench
- DiG-bench
- DirectorBench
- DisasterBench
- DisasterBench
- DiscoBench
- DiscoverPhysics
- DistilBERT ONNX Quantization Benchmark
- Distract-Bench
- DLawBench
- DME
- DMV-Bench
- DNBENCH
- Do LLMs Trust the Accuser or the Accusation? Measuring Belief Shifts in Werewolf
- Doc2CI
- Doc2DB-Bench
- DocFormBench
- DocHop
- DocPrivacyBench
- DocVQA
- DocVQA
- DocVQAtest
- Does Explainability Transfer? A Controlled Benchmark of Attribution Methods on Vision Transformers and CNNs
- DOSEBENCH
- Doubao Multi-Turn Bench
- DOVE
- Dr-CiK
- DRACO
- DRACO
- DragOn
- DRAGON
- DramaChain Bench
- DrawAI-Bench
- DrawingVQA
- DRFLOW
- DRGBT-1K
- DRInQ
- DriveMotion
- DRIVESPATIAL
- DROP
- DROP
- DROP
- DrugBench
- DS-1000
- DS-Arena-Code
- DS-FIM-Eval
- DSAgentBench
- DSBench-FullStack
- DSBench-Hard
- DSEffi-Bench
- DUDE
- DuMateBench
- DungeonBench
- DunphyBench
- DuoBench
- DVBench
- DyCodeEval
- Dyn-3D
- DynaBench
- DynaMath
- DynaMath
- DynamicMem
- E-Bench
- E-Commerce Bench
- E-EVAL
- EarlyDx
- Earnings25
- EarthShift
- EarthVerse
- EASEL
- EBench
- EC-Reason-Bench
- ECCBench
- ECG-InterpBench
- ECGQuest
- Echoverse
- ECLeKTic
- EcoAgent-Bench
- EComAgentBench
- EDGE-BENCH
- EdgeBench
- EdgeMosaic
- Edit2TikZ
- EditCLEVR
- EditInspector
- EduArt
- EduClaw-Bench
- EduPluginBench
- EduVideoBench
- EEG-EditBench
- Ego-MC-Bench
- Ego-METAS
- EgoAfford
- EgoBench
- EgoGapBench
- EgoGenEval
- EgoInertia-MI
- EgoMaize
- EgoMonth
- EgoNormia
- EgoProx
- EgoSafe-Bench
- EgoSafetyBench
- EgoSAT
- EgoSchema
- EGOSTREAM
- EgoTactile
- EHR-Complex
- EHR2Dial-Triage
- EHRBench
- EHRNote-ChatQA
- EHRNoteQA
- EIBench
- ELBench
- ELBench
- ElderBench
- ElephantBench
- Embodied-BenchForge
- Embodied3DBench
- EmbodiedAgentInterface
- EmbodiedBench
- EmbSpatialBench
- eMEM-Bench
- Emission-factor accuracy benchmark
- EMMA
- EMMA
- EmoSBench
- EMPATH
- Endo-C6
- EnergyBridge
- Enginuity
- ENPMR-Bench
- EnterpriseClawBench
- EnterpriseMem-Bench
- EnterpriseOps-Gym-AA
- EnterpriseRAG
- ENTLORE
- EntSQL
- EnvShip
- EpiBench
- EpiBench
- EPRSTMT
- EQ-Bench
- ER-Reason
- ERBench
- ERGeoBench
- ERP-Bench
- ERPBench
- ERQA
- ERQA-Plus
- ERUnderstand
- ESCUCHA
- ESF-Bench
- ESRP-Bench
- EuroExec
- European-IFEval
- euskal-agent-bench
- EvalAwareBench
- EvalLab
- EvalPlus
- Evaluation Agent
- Event ActivityNet
- EVID-Bench
- Evidence Package Benchmark
- Evo-Bench
- EvoAgentBench
- EvoArena
- EvoBrowseComp
- EvoClawBench
- EvoCode-Bench
- EvoGenUI-Bench
- EvoHarmBench
- EvoNav-Bench
- EvoPolicyGym
- EvoShift-Bench
- EWMBench
- EXE-Bench
- ExPhy
- ExplainBench
- Explanation Stability of Test-Time Adaptation in Computational Pathology: A Large-Scale Benchmark
- ExploitBench
- ExploitBench
- ExploitGym
- ExploitGym
- ExpoMotion
- ExtractBench
- ExtremeWhenBench
- EYT-Bench
- F-ICL
- FACTS Grounding
- FActScore
- FailBench
- Failure-Transparent Agents
- FairFund-Bench
- FAIRLENS
- FaithformBench
- FakeI2V-Bench
- FalseReject
- FAM-Bench
- FastKernels
- FaulT-Bench
- FBHM
- FEABench
- FedCMAPSS
- FedMABench
- FEPBench
- Fiducia-bench
- FigQA
- FilmBench
- Fin-Eva
- FinalityBench
- Finance Agent
- Finance Agent v1.1
- Finance Agent v2
- financial-sentiment-benchmark
- FinBalance
- FinBench
- FinBoardBench
- FinCUABuildBench
- FinDeepIndicator
- FinED-Bench
- FineSightBench
- FinEvo-Bench
- FinEvolveBench
- FinExam-10K
- FinFraudBench
- FinGuard-Bench
- FinHardBench
- FinixDocBench
- FinLifeBench
- FinPerMA
- FinPersona-Bench
- FinProBench
- FinQA
- FinRank
- FinRCA-Bench
- FinReportBench
- FinRiskAtlas
- FinSearchComp T2&T3
- FinSearchComp-T3
- FinStressTS
- FinVerBench
- FinVerse
- FIRM-Video-Bench
- FitAQA
- Flame-VLM-Code
- Flames
- FlameVQA
- FlatLab
- FlenQA
- FLEURS
- FLEX
- FLOATBench
- FloodReasonBench
- Flores
- FLUB
- FM-Bench
- FMG-Bench
- FolDeX
- FoodMonitor
- FORCE-Bench
- FORCEBENCH
- ForecastBench-Sim
- Forensics-bench
- ForeSci
- ForesightSafety-VLA
- ForestBench
- FORGE
- ForgetBench
- FormalTCS
- FormStruct-Bench
- FortisAVQA
- FPCO-Dialog
- FrameBench
- FRAMES
- FraudBench
- FREB-TQA
- FreightBidBench
- French MMLU
- FreshCache-Bench
- FriendBench
- From Financial Sentiment Classification to Return Predictability
- Frontier Agent Benchmark (FAB)
- Frontier Financial Judgement
- Frontier Science
- Frontier-Bench v0.1
- FrontierChallenge
- FrontierChallenge
- FrontierCode
- FrontierCode
- FrontierCode 1.1
- FrontierCS
- FrontierFinance
- FrontierMath
- FrontierMath Tier 4 (v2)
- FrontierScience Olympiad
- FrontierScience Research
- FrontierSWE
- FrontierSWE (Impl.)
- FullStackBench en
- FullStackBench zh
- FunctionalMATH
- FunPiQ
- FutureSurf
- FuzzingBrain-Bench
- G-IdiomAlign
- GABench
- GAIA
- GAIA2
- GAMA-Bench
- GameCraft-Bench
- GameEngineBench
- GameWorld
- GameXpert-Bench
- GAOKAO-Bench
- GAOKAO-MM
- GateMem
- GAUGE
- GAUGE
- GAUGE
- GauntletBench
- Gaze-Co
- GB/T-Bench
- GBA-GCs
- GBU-Palm
- GCA-Bench
- GDP.pdf
- GDP.pdf
- GDPevo
- GDPval
- GDPval
- GDPval-AA
- GDPval-AA v2 (Elo)
- GDPval-AA v2 (normalized score)
- GDPval-MM
- GDPval-Rubrics
- GEB-Bench
- GenAI-Bench
- GENEB
- GeneBench
- GeneBench-Pro
- General-Bench
- Generative Embedding Benchmark
- GenIaC-SecBench
- GenScale
- GEO-Bench
- GeoAgent
- GeoBenchLLM
- GeoDisaster
- GeoFidelity-Bench
- GEOID-Flood
- GeoNatureAgent Benchmark
- GeoStore
- GeoT2V-Bench
- GeoVAD-Bench
- GiantSteps Tempo
- GISAgentBench
- GIScholarBench
- GitGoodBench
- Global PIQA
- Global-MMLU
- Global-MMLU-Lite
- GlobalDentBench
- GlobeAudio
- GlucoFM-Bench
- GMAIMMBench
- Gorilla
- Gorilla Benchmark API Bench
- GovReport
- GPQA
- GPQA
- GPQA (full)
- GPQA Biology
- GPQA Chemistry
- GPQA Diamond
- GPQA Diamond
- GPQA Physics
- GPT-ImgEval
- GPTNT
- GPU-Accelerated ML Inference Benchmark
- GRACE
- GrailQA
- GranFact
- GraphARC
- GraphInfer-Bench
- GraphRareBench
- GraphVerse
- GraphWalks
- Graphwalks BFS 1M
- Graphwalks BFS <128k
- Graphwalks BFS >128k
- Graphwalks parents <128k
- Graphwalks parents >128k
- GreekBarRetrieval
- GRIT
- GroundingSuite
- Groundtruth Dynamic Benchmarking
- GroundUI-1K
- GroupToM-Bench
- GroupTravelBench
- Gsci-Fact
- GSM-8K (CoT)
- GSM1k
- GSM8K
- GSM8k
- GSM8K
- GSM8K Chat
- GSM8K-V
- GSO-Net
- GST-Bench
- GTA
- GTBench
- Guardian Crawler
- GuardianAgentBench
- GuardianBench
- GUI-CC
- GUI-Primitives
- GuideMe
- GUITestScape
- Gym4ReaL
- H2HMem
- H2R-Bench
- Hack-Verifiable Terminal Bench
- HAKARI-Bench
- HakushoBench
- Hallusion Bench
- HallusionBench
- HalluTruthQA
- HaluEval
- HamQASBench
- HANDBOOK.md
- Harbor-Index
- HARDMath2
- HardMTBench
- HarmURLBench
- HarmVideoBench
- Harness-Bench
- HarnessDev
- HarnessOpt-Bench
- HarnessRisk
- HarnessSafe
- Harvey LAB (Vals)
- Harvey LAB-AA
- Harvey LAB-AA
- HawkesNest
- HealMed
- HealthAgentBench
- HealthBench
- HealthBench
- HealthBench Consensus
- HealthBench Hard
- HealthBench Professional
- HealthBench Professional
- HealthBench-Psych
- HealthLoopQA
- HEART-Bench
- HeatCast
- Hedge-Bench
- HEDGEHOG
- HellaSwag
- HellaSwag
- HelloBench
- HERA
- HeraBench
- HERB
- HERO
- HERO'S JOURNEY
- HG-Bench
- HiddenMath
- HierSVA-B
- HighTide
- HighwayScene
- HiMed
- HindsightBench
- HiPhO
- HLE-Verified
- HLL
- HMMT
- HMMT 2025
- HMMT Feb 26
- HMMT25
- HoF-Bench
- HoliBench
- HOLMES
- HoloBench
- HoloCount
- HoloFair
- HoosierHelp
- HopRefusalBench
- HorizonMath
- HorizonMath
- HotpotQA
- HounsBench
- HouseCorr3D
- HR-Bench (4k)
- HRIBench
- HT-Bench
- HtFLlib
- HUG-VIS
- HumanCLAW-Bench
- HumanEval
- HumanEval
- HumanEval
- HumanEval Plus
- HumanEval+
- HumanEval-Average
- HumanEval-ER
- HumanEval-Mul
- HumanEval-X
- HumanEval-X++
- HumanEvalFIM-Average
- Humanity's Last Exam
- Humanity's Last Exam
- Humanity's Last Exam (no tools, text-only)
- Humanity's Last Exam (with tools, text-only)
- Humanity’s Last Exam
- HumanMoveVQA
- HumanoidArena
- HumanoidVLN
- HUSH-Bench
- Hy-MultiTurn
- HybridCodeAuthorship
- HybridDeepResearch
- HyperAlign-Bench
- HyperImageNet
- HyperShadow
- Hypersim
- HypoBench
- HypoEval
- I-WebGenBench
- IaC-Eval
- IBA-Bench
- ICAE-Bench
- IdeaAMBIG
- IdeaGene-Bench
- IDEAL-Bench
- IdioLink
- IDP-Bench
- IF
- IFBench
- IFBench
- IFBench
- IFCMemoryBench
- IFEval
- IFEval
- IFEval
- IFHierBench
- IFIR
- IFMTBench
- IH-Benchmark
- IHBench
- II-Bench
- Imag-Eval
- Image2FloorPlan
- ImageMining
- Imaging-101
- ImagingBench
- IMBench
- IMCBench
- IMDL-BenCo
- IMO 2025
- IMO-AnswerBench
- IMOAnswerBench
- IMOProof-Adv
- IMUG-Bench
- Include
- INCLUDE-BENCH
- independence-bench
- Indian Agent Benchmark
- Indic DiarBench
- IndicMMLU-Pro
- Industrial Dexterity Benchmark
- IndustryBench-MIPU
- InferQ
- InfiBench
- InfiniteBench/En.MC
- InfiniteBench/En.QA
- InfoBench
- InfographicsQA
- InfoOps Bench
- InfoVQA
- InfoVQAtest
- InfraBench
- InfraOcc
- INS-ActBench
- InSight
- INSPIRE
- Instruct HumanEval
- Instruction Stacking Collapse
- InstructMove
- InstruSum
- InsufficiencyBench
- Int-Bench
- IntegrityBench
- Intentbench-Prime
- IntentionNav
- Inter-X++
- InteractBench
- InterFLOPBench
- InterGPS
- Internal API instruction following (hard)
- Internal Research Debugging Evaluation
- InternData-A1
- InternData-M1
- InternData-N1
- InterveneBench
- IntPhys2
- Inverse IFEval
- Inverse Turing Bench
- InvestLogicBench
- InvestPhilBench
- iOSWorld
- IPhO 2025
- IPIBench
- IPO Finance Agent
- IQBench
- IR275K
- IRTS-ToolBench
- IS-Bench
- Ishigaki-IDS-Bench
- IslamicTurathBench
- IsoSci
- ISYV-Bench
- ITBench
- ITBench-AA
- ITPEval
- J1-Bench
- JailTrickBench
- JaleesBench
- JamBench
- Janus
- JarvisBench
- JarvisGUI
- JavaVulBench
- JEDI
- JETO-Bench
- JITOMA-Bench
- JL1-CC&QA
- JL1-CD
- JMed48k
- Job Bench
- JOB-Complex
- JobBench
- JointAVBench
- JOR-Bench
- Joyo Kanji Yomi Benchmark
- JudgeBench
- JudgmentBench
- JuryBench
- K-Bench
- K-BrowseComp
- K-FinHallu
- K-Sort-Arena
- K9-Bench
- KC-Bench
- Kernel Bench L3
- KernelBench Hard
- KernelGen 1P
- KernelGenBench
- KeyFrame-Compass
- KGI-Bench
- KhatianDoc
- KIDBench
- Kimi Claw 24/7 Bench
- Kimi Code Bench v2
- KINA
- KineBench
- KITAB-Bench
- KMMLU-Redux
- KnowLogic
- KOFFVQA
- KoNA
- KOPA-Bench
- KOR-Bench
- KORBench
- Kotlin Benchmark
- KSAFE-MM
- KVDiagnosis
- L-Eval
- L3Cube-IndicQuest v2
- LABBench2
- LabOSBench
- LADBench
- LAIT (Literary AI Translation)
- LakeQA
- LakeQuest
- LAMBADA
- LAMOST-DET
- LaMP-QA
- Land-use relevance benchmark
- LangChoiceBench
- LangNavBench
- LANTERN
- LaRA
- Last Translation Benchmark
- LatentMD
- LayerRAG-Bench
- LayerWiseBench
- LayoutBench
- LBPP (v2)
- LCS-Bench
- LCSTS
- LDU-Bench
- LeakageBench
- Legal Agent Benchmark
- Legal Agent Benchmark
- LegalCiteTrust
- LENS
- Lethe
- LEVANTE-bench
- LexAgentHallu
- LexIssue
- LexKairos
- LexRubric
- LH-AVLN
- LIBAD
- LIBERO-Recover Benchmark
- LIBERO-VIFO
- LibEvoBench
- LifePlanner
- LifeSciBench
- LigBench
- LilyBench
- Lingjing
- LINGOLY
- LingoQA
- LingT2I
- Lit2Test
- LitTraceQA
- Live Gurbani Tracking
- LiveBench
- LiveBench
- LiveBench
- LiveBench 20241125
- LiveBrowseComp
- LiveCodeBench
- LiveCodeBench
- LiveCodeBench
- LiveCodeBench
- LiveCodeBench
- LiveCodeBench Pro
- LiveCodeBench Pro
- LiveCodeBench v5
- LiveCodeBench v5 24.12-25.2
- LiveCodeBench v6
- LiveCodeBench(01-09)
- LiveHouse-TS
- LiveK12Bench
- LiveLongBench
- LiveMacroEval
- LiveMathBench
- LiveMathematicianBench
- LiveSports-3K
- LiveSQLBench
- LL-Bench
- LLAMIA-Bench
- LLaVA-Bench
- LLM Abstract Reasoning Benchmark
- LLM Benchmark Panel
- LLM Summarization Benchmark — Brazilian Portuguese
- LLM-BabyBench
- llm-finetuning-benchmark
- llm-free-benchmark
- LLM-SoccerArena
- LLM-SRBench
- LLM-Uncertainty-Bench
- LLMTabBench
- LLMThinkBench
- LLVM-Bench
- lm-evaluation-harness
- LMAct
- LMArena Text Leaderboard
- LOCA-Bench (256k)
- Local agent benchmark (tools + loop + coding)
- LOCOMO-CONV
- LogDx-CI
- LogiMed-RoB
- LogiQA-plus
- LogiQAv2-plus
- LOKI
- LoMeVQA
- Long-Horizon-Terminal-Bench
- LongAV-Compass
- LongBench
- LongBench
- LongBench v2
- LongCodeBench
- LongCounsel-8
- LongDocBench
- LongDS-Bench
- LongEarth-Bench
- LongEgoRefer
- LongFact
- LongFact Concepts
- LongFact Objects
- LongJudgeBench
- LongLSMDC
- LongMedBench
- LongRCA Bench
- LongText-Bench
- LongVALE
- LongVideoBench
- LongVideoBench
- LongVLBench
- LongVQUBench
- LongWebBench
- LongWoF-Bench
- LoopArena
- LoopNav
- LoopsBench
- Loopzero
- LoRFT
- LoSoNA
- Lost in Speech
- LPFQA
- LSAT
- LTMbenchmark
- LU-500
- LUNAR
- LV-Eval
- LVBench
- LëtzCross
- M$^3$ISR
- M$^3$R-Bench
- M-GATE: Multilingual Grammar, Accuracy in Translation, and Efficiency Benchmark for Large Language Models
- M2ABench
- M3-DuplexBench
- M3T
- MA-ProofBench
- MAC-Bench
- MacAgentBench
- MacArena
- MacroBench
- MADB
- MADBench
- MaDI-Bench
- MafiaScope
- MAG
- MaliciousSkillBench
- MalSkillBench
- MamaBench
- Management Consulting Tasks (Internal)
- Manager Coercion Benchmark
- Mandate Bench
- ManiGuard-Bench
- MapReason-OSM
- MariSat
- MaritimeBench
- MARS-Bench
- MasDrift
- MASH-Bench
- MASK
- MASK
- Massive-STEPS
- MastermindEval
- Mat-Pref
- MATH
- MATH
- MATH (CoT)
- MATH-500
- MATH-500
- Math-Vision Diagrams
- MathAdv
- MathArena Apex
- MathArena Apex 2025
- MathBench
- MathQA
- MathVerse
- MathVerse
- MathVerse-Mini
- MathVision
- MathVision
- MathVision
- MathVista
- MathVista
- MathVista
- MathVista-Mini
- MatPhaseBench
- MatReplace
- MAVERIX
- MAVOS-DD
- MAXIFE
- MBA-Bench
- MBench
- MBPP
- MBPP
- MBPP
- MBPP ++ base version
- MBPP EvalPlus
- MBPP EvalPlus (base)
- MBPP pass@1
- MBPP Plus
- MBPP+
- MC-CXR
- MCiteBench
- MCP Atlas
- MCP Atlas
- MCP-Mark
- MCP-Persona
- MCP-Universe
- MCPEvol-Bench
- MCPMark
- MCR-Bench
- MCR-Bench
- MD-VQA
- MDArena
- MeasureBench
- MECoBench
- Med-Fact
- MedAgents-Bench
- MEDAL
- MedArabiQ
- MedBench
- MedBench
- MedBookVQA
- MedBrowseComp
- MedCalc-Bench
- MedChemBench (Internal)
- MedClawBench
- MedCTA
- MedCUA-Bench
- MedDDC-Eval
- MedFailBench
- MedGym
- MedHal-Loc
- MedHallTune
- MedHallu
- MEDit-Bench
- MedJourney
- MEDLAYXPLAIN
- MedPIC-Bench
- MedPRESS
- MedRealMM
- MedReaMM
- MedSafetyBench
- MEDSEGBENCHMARKER
- MedStreamBench
- MedXpertQA
- MedXpertQA
- MedXpertQA-MM
- MeetingToM
- MEGA MLQA
- MEGA TyDi QA
- MEGA UDPOS
- MEGA XCOPA
- MEGA XStoryCloze
- Meld
- MemeBench
- MemeCULT-1K
- MemFail
- MemHop
- MemoBench
- MEMOBench
- MemOps
- MemoryDocDataSet
- MEMPROBE
- MemSecBench
- MemSyco-Bench
- MemTrace
- MeniOmni
- MEP-Bench
- MER-UniBench
- MER2026 MER-FG
- MergeMedBench
- MeshSplatBench
- Meta Internal Coding Bench
- MetaboNet-Bench
- MetaphorVU-Bench
- MetaRoute-Bench
- Metric-Bench Test
- MetroLLM-Bench
- MEventBench
- MEWC
- MGSM
- MHER
- MIABench
- MIB
- MicroVQA
- MIEB
- MiGUE-Bench
- MiLiC-Eval
- MIMIC CXR
- MiMo Coding Bench
- Mind the Gap: Robustness Risks in PII Detection Systems
- Mind2Web
- MindEdit-Bench
- MindTopo
- MineExplorer
- Minerva
- MineXplore
- miniCTX
- MiniF2F-ALF
- MiniLongBench
- MIRA-Ev
- MIRA-Math
- MiraBench
- MIRACL-VISION
- MIRAGE
- MirageBench
- MirrorCode
- MirrorCraft
- MissingBench-Verified
- MissionBench
- MJ-Bench
- MKG-RAG-Bench
- MKQA
- MLCR-AA
- MLE-Bench
- MLE-bench
- MLE-Bench Lite
- MLingualFC
- MLRC-Bench
- MLS-Bench Lite
- MLUBench
- MLVU
- MLVU
- MLVU-M
- MM IF-Eval
- MM-AlignBench
- MM-BrowserComp
- MM-ClawBench
- MM-CreativityBench
- MM-IFEval-Pro
- MM-IQ
- MM-Mind2Web
- MM-MT-Bench
- MM-RLHF
- MM-Snowball
- MM-ToolSandBox
- MM-Vet
- MMAD
- MMAE
- MMAR
- MMArch
- MMAU
- MMAU Music
- MMAU Sound
- MMAU Speech
- MMBC
- MMBench
- MMBench
- MMBench-Live
- MMBench-V1.1
- MMBench-Video
- MMBench-Video
- MMDG-Bench
- MMDS-Bench
- MMDU
- MME
- MME
- MME-CoT
- MME-RealWorld
- MME-RealWorld
- MMed-Bench-IR
- MMGait
- MMGist
- MMHBench
- MMIE
- MMIR
- MMIU
- MMJailBench
- MMKE-Bench
- MMLongBench
- MMLongBench-128K
- MMLongBench-Doc
- MMLongBench-Doc
- MMLongEmbed
- MMLU
- MMLU
- MMLU
- MMLU (CoT)
- MMLU Chat
- MMLU French
- MMLU-Base
- MMLU-Pro
- MMLU-Pro
- MMLU-Pro
- MMLU-ProX
- MMLU-Redux
- MMLU-Redux
- MMLU-redux-2.0
- MMLU-STEM
- MMMLU
- MMMLU
- MMMU
- MMMU
- MMMU
- MMMU (val)
- MMMU (validation)
- MMMU-Pro
- MMMU-Pro
- MMMU-Pro
- MMMU-Pro (with tools)
- MMMUval
- MMOOC
- MMPCBench
- MMPIBench
- mmPISA-bench
- MMS-VPR
- MMSearch
- MMSearch
- MMSearch-Plus
- MMShopBench
- MMSI-Bench
- MMSIBench
- MMStar
- MMStar
- MMT-Bench
- MMT-Bench
- MMTB
- MMTClinic
- MMVet
- MMVetGPT4Turbo
- MMVU
- MMVU
- mmWave-QA
- MNIST-PRO
- MobileForge
- MobileJudgeBench
- MobileMiniWob++_SR
- MobilePA-Bench
- MobileWorld
- MobileWorldSafety
- MoHallBench
- MOLE
- MolPuzzle
- MolSafeEval
- Moment-Video
- Momento
- Mono2Stereo
- MORE
- MORSE-500
- MortarBench
- MOSAIC
- MosaicLeaks
- MotionBench
- MotionBench
- MotionBlind
- MotionHalluc
- MotionMillion
- MOV-Bench
- Moving Alphabet
- MPAR-Bench
- MPC-Patch-Bench
- MPDocBench-Parse
- MPIE-Bench
- MR-Ben-Meta-Reasoning-Benchmark
- MR-GSM8K
- MR-LiDAR
- Mr.LHDR
- MRAG-Bench
- MRCR
- MRCR
- MRCR 128K (2-needle)
- MRCR 128K (4-needle)
- MRCR 128K (8-needle)
- MRCR 1M
- MRCR 1M (pointwise)
- MRCR 64K (2-needle)
- MRCR 64K (4-needle)
- MRCR 64K (8-needle)
- MRCR v2
- MRCR v2 (8-needle)
- MRCR v2 (8-needle, 512K-1M)
- MRMAD
- MS-MLB
- MS_MARCO
- MSE-Bench
- MSEval
- MSIBench
- MSP-Bench
- MSQA
- MSQA
- MSVE-Bench
- MT-AIME 2025
- MT-Bench
- MT-Bench-101
- MT-Web2Code
- MTAVG-Bench
- MTCMB
- MTEB
- MTEB-BR
- MTM-Bench
- MTOB
- MTPaperBananaBench
- MTVQA
- MTVQA
- MUDDLE
- MuirBench
- MulRobBench
- Multi-Challenge
- Multi-IF
- Multi-LCB
- Multi-Legal-Bench
- Multi-Organ CT Segmentation Benchmark
- Multi-SWE-Bench
- Multi-SWE-Bench
- Multi2AV-Safety
- Multi4D
- MultiChallenge
- MultiChallenge
- MultiGhostBench
- MultiGlobeQA
- MultiLF
- Multilingual MGSM (CoT)
- Multilingual MMLU
- MultiLoKo
- Multiparty Bench (MP-Bench)
- MultiPL-E
- Multipl-E HumanEval
- Multipl-E MBPP
- MultiRef-Compass
- MULTISEISMO
- MultivationBench
- MultiView-Bench
- MUSE
- MuseBench
- MusicCaps
- MuSP-Bench
- MuSR
- MusTBENCH
- MV-Bench
- MV2
- MVBench
- MVbench
- MVBench
- MVL-SIB
- MVPBench
- MyPCBench
- NAFBench
- NanoGPT
- NarrativeWorldBench
- NARU
- Natural Questions
- Natural2Code
- NaturalCodeBench
- NaturalProofs
- NatureBench
- NavVerse
- NBA_Streaming
- NCP-Bench
- NEEDL-Bench
- NEO-Bench
- Net-Ev$^2$-6.5M
- NetArtifactBench
- NetConfArena
- NetInjectBench
- NetlistBench
- Neurai-VN Benchmark
- NewsBench
- NextMotionQA
- Nexus
- NHANES Accelerometry Cardiometabolic Benchmark
- NIH/Multi-needle
- NL-PDDL-Bench
- NL2Repo
- NL2Repo
- NL2Scratch
- NL2SHACL-Bench
- NLP Sentiment Analysis Benchmark
- NMOS
- nolima
- NoLiMa 128K
- NoLiMa 32K
- NoLiMa 64K
- NOLLI
- NoRA
- NormAct
- NormBench
- NOVA-63
- NovelAPIBench
- NovGauge
- NPPC
- NQ
- NQ
- NQP-Bench
- NRITYAM
- NRT-Bench
- NuclearQAv2
- NumBench
- NumerosityVLM
- Nuscene
- NutriBench-Kitchen
- NutritionQA
- O.W.L. & N.E.W.T. Bench
- OAO-AttackBench
- ObGynLongBench
- Objectron
- OBLIVION
- ObsDriveBench
- ObserverBench
- Obshazard-bench
- OCCUR-Bench
- OCNLI
- OCR-MetaReasoning
- OCRBench
- OCRBench
- OCRBench-V2 (en)
- OCRBench-V2 (zh)
- OCRBench_V2
- OCT-Bench
- OctoCodingBench
- OdinEval
- ODinW
- OEIS Open
- OenoBench
- Office Comprehension Benchmark
- OfficeQA Pro
- OfficeQA Pro
- OffNadirLoc
- OJBench
- OJBench (C++)
- OlymMATH
- OlympiadBench
- OlympiadBench
- OlympicArena
- OmegaUse-OfficeVal
- OmissionBench
- Omni-MATH
- OmniaBench
- OmniAlign-V
- OmniBench
- OmniBench
- OmniBench Music
- OmniCAD
- OmniCap-IF
- OmniDocBench
- OmniDocBench
- OmniDocBench
- OmniDocBench 1.5
- OmniEdit-Bench
- OmniEEG-Bench
- OmniFood-Bench
- OmniGAIA
- OmniGameArena
- OmniGIRL
- OmniHandwritingOCR
- OmniInteract
- OmniLayout
- OmniMapBench
- OmniMatBench
- OmniMath
- OmniMech
- OmniMMI
- OmniOpt
- OmniPhys
- OmniPhys
- OmniRef-Bench
- OmniRetriever-Bench
- OmniRouting
- OmniScience
- OmniScience (non-hallucination rate)
- OmniSmall
- OmniToM
- OmniTraffic
- OmniVideoBench
- OMTG Bench
- OncoTraj
- OncoTriad-QA
- One Word, Different Action
- ONECYL
- OneMillion Bench
- OneMillionBench
- onepot-Bench
- OntologyBench
- OOD-RL-Bench
- OpAI-Bench
- Open-rewrite
- OpenAI MMLU
- OpenAI-MRCR: 2 needle 128k
- OpenAI-MRCR: 2 needle 1M
- OpenAI-MRCR: 2 needle 256k
- OpenBenchmark
- OpenBookQA
- OpenbookQA
- OpenFinData
- OpenHalDet
- OpenHarmony Bench
- OpenRCA
- OpenRTAG
- OpenSafeIntent
- OpenSciToolBench
- OpenSkillRisk
- OpenTuringBench
- OpenUnlearning
- OPT-BENCH
- Opti-Agent-Bench
- OptimBench
- OR-Bench
- OR-Space
- OraclePhys
- ORAgentBench
- Orak
- ORCA-bench
- OrchBench
- OSGuard
- OSS-Bench
- OSWorld
- OSWorld
- OSWorld 2.0
- OSWorld 2.0
- OSWorld Extended
- OSWorld Screenshot-only
- OSWorld-G
- OSWorld-Verified
- OV-SGG-Bench
- OV3D-Bench
- OVBench
- OVEarth-Bench
- overlooked-bench
- OVIBench
- OVO-S-Bench
- OVOBench
- P-MMEval
- P3B3
- P3D-Bench
- PACE-Bench
- PACUTE
- PaintBench
- PAIR-Bench
- Pak3H
- PAL-Bench
- Pancasila-Dilemmas
- PaperBench
- PaperBench
- ParaGUIBench
- ParamBench
- ParaPairAudioBench
- ParaRecover
- PARARULE-Plus
- ParasGB
- Particle Event Classification with Classical and Hybrid Quantum Models: A Reproducible Benchmark
- PaSBench-Video
- PashtoOCR
- PashtoTTS-Bench
- PAST-Bench
- PatchBench
- PatchFuseBench
- PathAgentBench
- PathMCQA
- PathoArgus-Bench
- PathReportEval
- PathView-Bench
- PatientAgentBench
- PatternEval
- PAUSE
- PAWBench
- PCA-Bench
- PCBnet
- PeakBench
- PEFT-Arena
- Peg-in-Bench
- Pelican Bike Benchmark
- Peony
- PerceptionBench
- PerceptionBench
- PerceptionTest
- PereStruct benchmark
- PERFOPT-Bench
- PerMemBench
- PersLitEval
- PersonalBench
- PersonaLens
- PersonaShot
- PersonaTrail
- PerspectiveGap
- PertEval-scFM
- PetQA: Benchmarking Veterinary Knowledge and Clinical Reasoning
- PhAIL (Physical AI Leaderboard)
- PhantomBench
- PhantomFill
- PhenoNormBench
- PhiBench
- PHITSBench
- PhoneHarness
- PHREEQC-MCQ-200
- Phun-Bench
- PHYBench
- PhyBench
- PhyEditBench
- PhyGenBench
- PhysAssistBench
- PhysElite
- Physics-IQ Verified
- PhysicsBench
- PhysicsFinals
- PhysMani-Bench
- PhysReason
- PhysTool-Bench
- PICABench
- PinchBench
- pinn-gym
- PInVerify
- PIPBench
- PIPE-Cypher
- Pipette
- PIQA
- PIQA
- PitchBench
- PitfallBench
- PIVOT
- PIVOTSBench
- PL-NBA
- Plan2Map
- PlanBench
- PlanBench-V
- PlanBench-XL
- PLawBench
- PlayWorld
- PLCBENCH
- PLSQLBench
- PluginEval
- PM-Bench
- PMC-VQA
- PNI-3D Multimodal CAD & Geometric Robustness Benchmark
- Pocket-Dentist
- POINav-Bench
- PointGrounding
- PointQ-Bench
- PoisonForge
- PokerBench
- PolicyShiftBench
- Polistemics
- PolyComp
- PolyMATH
- PolyMath-en
- PolySpeech-100
- PolyWorkBench
- PoolBench
- POPE
- POPE
- PopMCQ
- PopQA
- POPSICLE
- PortBench
- PorTEXTO
- Porting Benchmark
- Poryos2026
- PosterBench
- POSTERHARNESS
- PosterSum
- PostTrainBench
- PostTrainBench
- PostTrainBench Lite
- Power Systems Agent Benchmark
- PowerCodeBench
- PPE-Bench
- PPT-Eval
- PRAGMA
- PragMatch
- PRBench-Finance
- PRBench-Legal
- Pre-Flight
- PreAct-Bench
- Precision Is Not Faithfulness: Coverage-Aware Evaluation of Grounded Generation with a Complete Oracle
- PredAct-Bench
- PresentBench
- Principia
- Principle-Bench
- PrionNER
- PRISM
- PRISM-Bench
- PrivacyPeek
- PrivBench
- PriVE-Bench
- PRM-as-a-Judge 1.5
- PRMBench_Preview
- PRMU
- ProactiveBench
- ProBench
- ProcArena
- ProcessBench
- ProEvent
- ProfBench
- ProForma-20Q
- Program Bench
- ProgramBench
- ProjectionBench
- ProJudge
- Prompting-MammAlps
- ProofNet
- ProtocolQA
- ProtStructQA
- PRWeaver
- PseudoBench
- PTCG-Bench
- PTXBench
- PuMVR
- PyGeoX-Bench
- PyraMathBench
- Q-Bench
- Q2D-Web
- QASC
- Qasper
- Qiskit QuantumKatas
- QMSum
- QO-Bench
- QuechuaTok
- QuoteBench
- QVal
- QVHighlights
- Qwen-Image-Bench
- qwen3.8-27b-inference-benchmark-4090
- QwenClawBench
- QwenQoderBench
- QwenReactBench
- QwenSVG
- QwenSWEBench
- QwenWebBench
- QwenWorldBench
- R2HandoverSim
- R2M-Bench
- R3-Bench
- RA-Bench
- RACE(High)
- RACE(Middle)
- rag-retrieval-benchmark
- RAG-Safety-Bench
- RandomBench
- RATIO
- RDB2G-Bench
- RDVSv2
- RE-Bench
- ReactBench
- ReactHuman
- ReactSim-Bench
- READI
- REAL
- REAL-Bench
- RealBench
- RealCADBench
- RealDESED
- RealDocBench
- RealisticTritonBench
- RealKIE-FCC
- RealMath-Eval
- RealOOB
- RealToxicityPrompts
- RealWorldQA
- RealworldQA
- ReasonMatch-Bench
- RECAP
- Receipt Replay OOD
- ReceiptBench
- ReClor-plus
- RECOM
- Reconstruction
- ReCoRD
- recoveriq
- RecreationBench
- REDACT
- RedactionBench
- REDAgentBench
- RedCode
- RedundancyBench
- RedVox
- RefCOCO-avg
- RefCOCOg
- ReflexBench
- RefMem-Bench
- RefSpatialBench
- refusal-benchmark
- RegionDet
- RegretBench
- REKEY
- Relational Anomaly Detection Benchmark
- Relay-Bench
- REMEDI
- ReMemBench
- REPAIR-Bench
- RepBench
- RepLiQA
- Repo Env
- repo-context-ambiguity
- RepoBench
- RepoMirage
- RepoPeftBench
- RepoProbe
- RepoQA
- RepoReasoner
- ReRef-3D
- RESCAST-100K
- RESCUE
- RescueBench
- Research Attention Prediction (RAP)
- ResearchClawBench
- ResearchClawBench
- ResearchQA
- RESPClinBench
- RestoreBench
- RetailBench
- RevalExo
- Reveal
- RevengeBench
- ReViCo
- RewardBench
- RExBench
- RFUAV
- RIDE
- RIFT-Bench
- RIG-BENCH
- RigidBench
- RigorBench
- RigorousBench
- RISEBench
- RM-Bench
- RNG-Bench
- RoboDojo
- RoboMME-Interference
- RoboProcessBench
- RoboReel
- ROBORMBENCH
- RoboSemanticBench
- RoboSPA
- RoboSpatialHome
- RoboTrustBench
- RobotValues
- RoboVista
- RoboWits
- Robust IF
- RobustMAD
- RoleCDE
- RoleLLM
- Roman Numeral Harmony Benchmark
- RotOutBench
- ROUTEBENCH
- RPCBench
- RQ-Bench
- RRS-10K
- RS-RIE-Bench
- RSI Index
- RSI-Bench
- RSIBench-Data
- RSMMVP
- RTCbench
- RTE
- RTL-Bench
- RuBench
- RuleMaze
- RULER
- RULER 1000K
- RULER 128k
- RULER 2048K
- RULER 512K
- RULER 64k
- RuleShift-Bench
- RuleWeaver
- RuleWorld
- RUListening
- RusFinChain
- Rushes
- Russian IT Community Corpus
- RUT-Bench
- RuVerBench
- RW-Voice-EQ Bench
- RWGBench
- RXRX3-CORE
- S-Eval
- S1-Bench
- S2M-Bench
- S2R-Bench
- S3Gym
- SA-Bench
- SABER
- SABER-Math
- SAEBench
- SAEScientist-Bench
- SafeAtlas-Bench
- SafeBuild-Bench
- SafeClawArena
- SafeClawBench
- SafeGen-Bench
- SafeGesture
- SafePyramid
- SafeRelBench
- SafeRun
- SafeSceneReason
- SafetyBench
- SafeVid
- SafeVLA-Bench
- SAFIRE
- SagaQA
- SAKE
- SALAD-Bench
- SalArt-VQA
- Samuel_Collins_CV_Benchmarking
- SAT Math
- SBMLLM-Bench
- SC3
- SCALE-QA
- SCAM
- scBench-Long
- SceneActBench
- SCHEDBench
- SchemaGUI
- SchemeArena
- Sci-MMR
- Sci-VBench
- SciAgentArena
- SciClaimEval
- SciCode
- SciCode
- SciCode
- SciConBench
- SciDocBench
- SciDraw-Bench
- Science Edge Evaluation
- ScienceArena
- ScienceQA
- ScienceQA
- ScienceQA Visual
- SciExplore
- SciFIBench
- SciFigBench
- SciFigPlag-Bench
- SciFigQual-Bench
- SciFigure2Code
- SciHazard
- SciIntBench
- SciIR-Bench
- SCILAWS-BENCH
- SciMIF
- SciR
- SciRisk-Bench
- SciStyleBench
- SCOPE-Bench
- SCR-Bench
- ScrambleToolBench
- ScratchWorld
- ScreenSpot
- ScreenSpot Pro
- scTranslation
- SDABench
- SDARE-Bench
- SDGBiasBench
- SDR-Bench
- SEA-SpeechBench
- Seal-0
- SEAM-Bench
- Search and Function-Calling
- SearchAuditBench
- SEC-bench
- SEC-bench
- SEC-bench Pro
- SEC-Bench Pro
- SecBench
- SecCodeBench
- SecCodeBench
- SecDrift
- SecRespond
- SEED-Bench
- SEED-Bench-2
- SEED-Bench-2-Plus
- SeedClawBench
- SeGaBench
- SelectBench
- SemCityLockeD
- SemCog Bench
- SemiMat
- SemVerBench
- SENSE-VAD
- Sentinel Challenge
- SentinelBench
- Seq2Synth
- SeqAlign3DVG
- Setoka
- SEVRA-BENCH
- SFE
- SG-Bench
- SGI-Bench
- SGR-Bench
- ShadowBench
- ShallowBench
- Shell
- Shift & Drift
- Shopping Reasoning Bench
- ShoppingMMLU
- SHOVIR
- SidewalkBench
- SIFO
- SIFO-Multiturn
- SIGNPOST-Bench
- Silent Misresolution
- silentbug
- Sim2Real-AlgoBench
- Sim2Signal
- SIMMER
- SIMPLE
- SimpleQA
- SimpleQA
- SimpleQA Verified
- SimpleVQA
- SingGuard-Bench
- Single-answer Atomic Long-form Target (SALT)
- SIQA
- Siren AgentDojo Attack Success Rate
- Siren AgentDojo Utility
- SIREN-Bench-v1
- SIS-Bench
- Skill-Use
- SkillAdeqBench
- SkillChain-Gym
- SkillEvolBench
- SkillHarm
- SkillMisevo-Bench
- SkillReason-Bench
- SkillResolve-Bench
- SkillSafe-Bench
- SkillsBench
- SkillsBench
- SkillTV-Bench
- SkillVetBench
- SkillVetBench
- SkySeaLand
- SlakeVQA
- SLAPBench
- SLVMBench
- SmartBench
- SmellBench
- SMH-Bench
- SMMILE
- SnapBench
- Social IQa
- SocialPersona
- SocietyBench
- SOCO
- SoCRATES
- SocSci-Repro-Bench
- SoftVTBench
- SoftVTBench: A Deformation-Aware Visuo-Tactile Dataset and Benchmark for Deformable-Object Manipulation
- SolarChain-Eval
- SolidityBench
- SoMBench
- SopriBench
- SoundnessBench
- SovereignNegotiation-Bench
- SovereignPA-Bench
- SP-Bench
- SP-TransientBench
- SPACE
- SPADE
- SPADE-Bench
- SpaPath-Bench
- SPAR-Bench
- Spatial457
- SpatialBench
- SpatialBench-Long
- SpatialGen-Bench
- SPATIALQUERY-1M
- SpatialUAV
- SpatialWorld
- SPEARBench
- SpeechDx
- Spider
- Spider
- Spider 2.0-AIFunc
- Spider2-V
- SPIEval
- SPIKE-Bench
- SPINE
- SportMV-Bench
- SportQA
- SPOT-THE-SHIFT
- Spreading-Oriented Reduction Benchmark (SORB)
- SpreadsheetBench
- SpreadsheetBench
- SpreadsheetBench 2
- SpreadSheetBench-v1
- SQBench
- SQuALITY
- SRE-Bench
- SSMNBench
- ST-Bench
- ST-Evidence
- StabilityBench
- StableToolBench
- STAGE-Claw
- StakeBench
- StaminaBench
- StanceBench
- StanceFlip
- STARK_10k
- StartupBench
- StatABench
- Statebench
- StatFormBench
- StatMechBench-v0
- StealthBench
- steam-review-sentiment
- SteelBench
- SteerBench
- SteerBench-Work
- STEM
- StemBind
- STEMGym
- StepJack
- StereoGenBench
- StocBench
- StochBench
- STORYLENSBENCH
- StrAD
- strategy-benchmark
- StrategyBench
- StrategyQA
- StreamArena
- StreamMemBench
- StreamProfileBench
- StructFlowBench
- StructTokenBench
- StudentEval
- StudyBench
- STVG-CompBench
- StyleRec
- StylisticBias
- SubtleMemory
- SuiChat-CN
- SummScreenFD
- SUNRGBD
- SuperChem
- SUPERChem
- SuperGLUE
- SuperGPQA
- SuperGPQA
- SuperGPQA
- SurakshaEval
- SURDS
- SurgVLA-Bench
- SurgWMBench
- SurveyReview
- SVAMP
- SVG-Bench
- SVGEval
- SVHalluc
- SVI-Bench
- SWE Atlas
- SWE Atlas - Codebase QnA
- SWE Atlas - Test Writing
- SWE Refactor Bench
- SWE-Atlas
- SWE-bench Multilingual
- SWE-bench Multilingual
- SWE-Bench Multimodal
- SWE-Bench Pro
- SWE-bench Pro
- SWE-Bench Pro Verified
- SWE-Bench ProMax
- SWE-bench Science
- SWE-bench Science
- SWE-Bench Verified
- SWE-bench Verified
- SWE-bench Verified (Agentic Coding)
- SWE-bench Verified (Agentless)
- SWE-bench Verified (Multiple Attempts)
- SWE-bench-Live
- SWE-Explore
- SWE-Factory
- SWE-fficiency
- SWE-Gate
- SWE-InfraBench
- SWE-INTERACT
- SWE-Lancer
- SWE-Lancer
- SWE-Lancer (IC-Diamond subset)
- SWE-Marathon
- SWE-Marathon
- SWE-Marathon
- SWE-MM
- SWE-Mutation
- SWE-NFI
- SWE-Perf
- SWE-Review
- SWE-RPG
- SWE-Together
- SwiLTra-Bench
- SwissCrop25
- SWORD
- SWT-Bench
- SymbalBench
- SynCred-Bench
- SynthAVE
- SynthDocBench
- Synthetic Data Generator Benchmark (Credit Risk)
- T-Eval
- T-IMPACT
- T1-Bench
- t2-bench
- T2D-Bench
- T2J-Bench
- T2LSC-Bench
- TabBench-Bio
- Tabero
- TabFact
- TableEval
- TabQueryBench
- TABVERSE
- TactiDex
- TacVerse
- TadA-Bench
- TAF-MED
- Tailor-Bench
- TailorBench
- TAKE 85
- TalkFa
- TangPoetryBench
- TAPVid-MV
- TAR-Bench
- TaskBench
- TAU-Bench
- Tau-bench
- tau-bench
- TAU-bench Airline
- TAU-bench Retail
- Tau2 Airline
- Tau2 Retail
- Tau2 Telecom
- tau2-bench
- Tau3 Airline
- Tau3 Banking
- Tau3 Retail
- Tau3 Telecom
- TAU3-Bench
- TC-Bench
- TCA-Bench
- TCR-Bench
- TCS-BENCH
- Teaching Monster Challenge
- TELBench
- TelemetrySuffBench
- TeleSWEBench
- TempCloze
- TempCompass
- Tencent WorkBuddy Bench
- TensorBench
- Terminal-Bench
- Terminal-Bench
- Terminal-Bench 2.0
- Terminal-Bench 2.1
- Terminal-Bench 3.0
- Terminal-Bench Hard
- Terminal-Bench Hard
- Terminal-Bench v2.1
- Terminus
- TerraBench
- TerraLogic
- TestEvo-Bench
- testing
- TeXFix-Bench
- Text2World
- TextRich
- TextVQA
- TF-RefusalBench
- TGLG
- Thai_local_benchmark
- The Complexity Ceiling Benchmark
- The Energy Society
- The Image Reconstruction Game
- The Imitator Game
- The Value Engine Benchmark (VEB)
- TheoremBench
- TheoremQA
- TheoremQA
- TherapeuticsBench
- Thinkingbox-Bench
- ThorArena
- ThousandWorlds
- Threat-Signature_Eval
- THUNDER
- TIC-Bench
- TickingCollabBench
- TIDE
- TIDE-Bench
- TIDE: A Physically Diverse 3D Turbulence Benchmark Dataset for Advancing Scientific Machine Learning
- TIER
- Time-Aware Multi-View MRI Benchmark
- TimeConflict
- TimeSage-EV
- TimeSage-MT
- TimeTravel
- TimeVista
- tiny_qa_benchmark_pp
- TIR-Bench
- TLA+-Bench
- TLDR9+ (test)
- TomaBench
- TOMATO
- Tool Decathlon
- ToolAlignBench
- Toolathlon
- Toolathlon Verified
- ToolBench-X
- ToolFailBench
- ToolHop
- ToolMaze
- ToolMenuBench
- ToolRet
- ToolRobustBench
- TouchSafeBench
- ToxScreen
- TRACE
- TRACE-Bench
- TraceJudgeBench
- TradeVerse
- Trae Code Gen
- Trae Error Fix
- TrajErrBench
- TransBench
- TransClean
- Translation en→Set1 COMET22
- Translation en→Set1 spBleu
- Translation Set1→en COMET22
- Translation Set1→en spBleu
- TransLaw
- TRAPSBench
- TravelEval
- TREAT
- TreeBench
- TreeProbe
- TREK
- TriggerBench
- TRINITY
- TriState-Bench
- TriviaQA
- TriviaQA
- TriViewBench
- TRL-Bench
- TrustDABench
- TruthfulQA
- TruthfulQA
- TruthInsightBench
- TS-Fault
- TS-Skill
- TSAI-MetaFraud
- TSC-Bench
- TSM-Bench
- TsuGO
- TSWAP Thai Wellness Benchmark
- TUA-Bench
- TukaBench
- Turkish LLM Benchmark
- TurnBench
- TurtleAI
- TVBench
- TVRBench
- tw-bench
- tw-exam-bench
- TydiQA
- TyDiQA
- U-NIAH
- UA-Legal-Bench
- UAV-OVO
- UAV3DCrop
- ubuntu_osworld
- UCF101-AD
- UESF-Bench
- UFPR-PEs
- UHGEval
- UHI-Bench
- UI2App
- UltraVR
- UMI-Bench
- Unconstrained Vehicle Identification Benchmark
- UnderSpecBench
- Uni-MMMU
- UniBench
- UniCAD
- UniClawBench
- Uniform Bar Exam
- UniHall
- UniKE
- UniQL
- Unison
- UnpredictaBench
- UOJ-Bench
- UpliftBench
- Urban Driving-Face Scene Gaze (UD-FSG)
- UrbanVideo-Bench
- UrbanWell
- UrduMMLU
- USAMO 2026
- USAMO25
- UserToolBench
- UTBoost
- UTILMEM
- UXBench
- UXBench
- V*
- V-STaR
- V2V-Bench
- VA-EvoBench
- VA-Judger-Bench
- VAKRA
- VakyArth
- VAmoS Bench
- VanillaBench
- VANTAGE-BENCH
- VARM-Bench
- Vast Benchmarking
- VATEX
- VBench
- VBench
- VBench++
- VBench-2.0
- VBVR-Pro
- VCIFBench
- VCR_en_easy
- Vector-Bench
- VEHBench
- Vending Bench
- Vending-Bench 2
- VendorBench-100
- Vera-Bench
- VeriOCRBench
- VeriTrip
- Vero
- VertiCue-Bench
- Verus-SpecBench
- VES-Bench
- VEX-Bench
- VG-GUIBench
- VGA-BenchV2
- VGenST-Bench
- VGI-BENCH
- VHDLBench
- VI-Bench
- VIABench
- VIABLE
- VialectBench
- VIBE
- VIBE
- VIBE
- VIBE Android
- VIBE Backend
- VIBE iOS
- VIBE Simulation
- VIBE Web
- VIBE-Bench
- Vibe-Eval
- VIBE-Pro
- VIBE-V2
- VibeLifeBench
- ViBench
- Video-IFBench
- Video-MME
- Video-MME
- Video-MME
- Video-MME (long, no subtitles)
- Video-MME-Logical
- VideoArgus-Bench
- VideoFDB
- VideoGAIA
- VideoGUI
- VideoHolmes
- VideoMathQA
- VideoMME w sub.
- VideoMME w/o sub.
- VideoMMMU
- VideoOdyssey
- VideoReasonBench
- VideoRover-Bench
- VideoSimpleQA
- VideoVIBE
- VideoWeaver
- VidHalLoc
- VidMsg
- VietFashion
- ViLBench
- VIPER
- VIPS
- ViroBench
- Virology Capabilities Test
- VisAnomBench
- VISCO
- VisEditBench
- VisFactor
- Vishing-Tactics-Bench
- Vision2Web
- VISTA (VIsual Spec-To-App Benchmark)
- VistaHop
- VisTarget-Bench
- ViStoryBench
- ViSTR-Bench
- Visual RAG Benchmark
- VisualFLIP
- VisualLeakBench
- VisualPuzzles
- VisualSimpleQA
- VisualWebBench
- VisuLogic
- VITA-Bench
- VitaBench
- VKnowU
- VLADBench
- VLM2-Bench
- VLMsAreBiased
- VLMsAreBlind
- VMetaphor-Bench
- VNISH GLOBAL Multilingual Operator Routing Benchmark
- VocalSound
- VoiceBench Avg
- VoxSumm
- VQA-Rad
- VQABench
- VQAv2
- VQAv2 (test)
- VQAv2 (val)
- VRBench
- VRL-Bench
- VSRo-200
- VTC-Bench
- Vul4Py
- VULBENCH-CPP
- VVA-Bench
- VVM-Bench
- VWG-Bench
- WADE
- WaspMOT
- WatchAct
- WaveformQA
- WBench
- WC2026-Agents
- We-Math
- WearableQA
- WeaveBench
- Web Bench
- WebArena-Verified
- WebDev Arena
- WebDev-Skills-Bench
- WeBench-MMGenEdit
- WebIGBench
- WebRetriever
- WebUI-Bench
- WebVoyager
- WeClawArena
- WeGenBench
- WeSCE
- WhatIfBench
- WhisperBench
- Who&When Pro
- WhodunitBench
- WiC
- WideSearch
- WideSearch
- WikiContradict
- WikiSQL
- Wild Bench
- WildBench
- WildBox
- WildClawBench
- WildClawBench
- WildHandBench
- WildTrace
- Winogrande
- WinoGrande
- WireCraft
- WMDP
- WMT23
- WMT24++
- WorkBench
- Workflow-GYM
- Workspace Bench
- WorkspaceBench
- WorkSurface-Bench
- WorldBench
- WorldBench
- WorldBench
- WorldCoder-Bench
- WorldCupArena
- WorldExam
- WorldGenBench
- WorldLines
- WorldOlympiad
- WorldRoamBench
- WorldScore
- WorldVQA
- WritingBench
- WritingBench
- WSADBench
- WSC
- WSE-bench
- WUICC-bench
- WuYu-EnvLE-Bench
- WuYuEval
- X+Slides
- X-Stream
- xCodeEval
- xDailyBench
- xDailyBench
- XEWorld
- XIH-Bench
- XL-DocBench
- XLSum English
- XPlainVerse
- XREPOTEST
- XSTest
- XSum
- xVerify
- YallaMorph
- YC-Bench
- YoCausal
- YOMI-Bench
- Yue_Benchmark
- ZClawBench
- ZDPShift
- ZebraLogic
- ZEROBench
- ZeroBench
- ZeroBench
- ZEROBench-Sub
- ZeroSight
- ZIPBench
- Zipbench
- π-SUB
- τ-Rec
- τ²-Bench Telecom
- τ³-Banking
- Φ-Bench