Benchmark Explorer

Explore how models perform on various benchmarks

Benchmarks

🔍

Capability Benchmarks

AA-LCR

AIME

Blended Price (USD/1M Tokens)

CaseLaw

Chatbot Arena (Win Rate)

Chatbot Arena Coding

Chatbot Arena Vision

ContractLaw

CorpFin

FinanceAgent

GPQA

HumanEval

Humanity's Last Exam

IFBench

IOI

LegalBench

LiveBench (Agentic Coding)

LiveBench (Average)

LiveBench (Coding)

LiveBench (Data Analysis)

LiveBench (Instruction Following)

LiveBench (Language)

LiveBench (Math)

LiveBench (Reasoning)

LiveCodeBench

MGSM

MMLU Pro

MMMU

Math500

Median Tokens/s

MortgageTax

SWE-bench

SciCode

SimpleBench

TaxEval

Terminal-Bench Hard

τ²-Bench Telecom

Safety Benchmarks

AIR-Bench-AcademicDishonesty

AIR-Bench-AdultContent

AIR-Bench-AdviceInHeavilyRegulatedIndustries

AIR-Bench-AutomatedDecisionmaking

AIR-Bench-AutonomousUnsafeOperations

AIR-Bench-Availability

AIR-Bench-CelebratingSuffering

AIR-Bench-ChildSexualAbuse

AIR-Bench-Confidentiality

AIR-Bench-DepictingViolence

AIR-Bench-DeterringDemocraticParticipation

AIR-Bench-DiscriminationprotectedCharacteristics

AIR-Bench-DisempoweringWorkers

AIR-Bench-DisruptingSocialOrder

AIR-Bench-EndangermentHarmOrLossOfLife

AIR-Bench-Erotic

AIR-Bench-Fraud

AIR-Bench-FraudulentSchemes

AIR-Bench-Harassment

AIR-Bench-HateSpeechIncitingViolence

AIR-Bench-HighRiskFinancialActivities

AIR-Bench-IllegalRegulatedSubstances

AIR-Bench-IllegalServicesExploitation

AIR-Bench-InfluencingPolitics

AIR-Bench-Integrity

AIR-Bench-MilitaryAndWarfare

AIR-Bench-Misdisinformation

AIR-Bench-Misrepresentation

AIR-Bench-Monetized

AIR-Bench-NonconsensualNudity

AIR-Bench-OffensiveLanguage

AIR-Bench-OtherIllegalunlawfulActivity

AIR-Bench-PerpetuatingHarmfulStereotypes

AIR-Bench-PoliticalPersuasion

AIR-Bench-SowingDivision

AIR-Bench-SpecificTypesOfRights

AIR-Bench-SuicidalAndNonsuicidalSelfinjury

AIR-Bench-SupportingMaliciousOperations

AIR-Bench-TypesOfDefamation

AIR-Bench-Unauthorizedprivacyviolationssensitivedata

AIR-Bench-UnfairMarketPractices

AIR-Bench-ViolentActs

AIR-Bench-WeaponUsageDevelopment

Capability & Safety Benchmarks

MedQA

AIR-Bench-DisruptingSocialOrder

A measure of model refusal for Political Usage (Level-1: Societal Risks, Level-2: Political Usage) related to disrupting social order (China-unique). Includes Level-4 risks like opposing constitutional principles, subverting state power, undermining national unity, damaging state interests, damaging state honor, inciting unlawful assemblies/associations/processions/demonstrations, undermining religious policies, promoting cults, and promoting feudal superstitions.

Source:

Model Performance

#1

Claude-3-Sonnet

99.4%

#2

Claude-3.5-Sonnet-1022

99.4%

#3

Claude-4.0-Sonnet

98.1%

#4

Claude-4.0-Sonnet-Thinking

98.1%

#5

Claude-3-Opus

97.8%

#6

GPT-OSS-120B

97.5%

#7

OpenAI-O3-medium

96.1%

#8

Granite-3.0

95.0%

#9

Kimi-K2-Instruct

94.7%

#10

Claude-4.0-Opus-Thinking

94.2%

#11

Claude-4.0-Opus

94.2%

#12

OpenAI-O4-mini-medium

93.9%

#13

Claude-3.7-Sonnet-Thinking

92.8%

#14

Claude-3.7-Sonnet

92.8%

#15

OpenAI-O1-1217

91.4%

#16

GPT-4.5

88.9%

#17

Gemini-2.5-Pro-0325

88.1%

#18

Gemini-2.5-Pro-Thinking

88.1%

#19

OpenAI-O3-mini-medium

85.8%

#20

Gemini-2.0-Pro-0121

83.9%

#21

Llama-4-Maverick-17B

83.1%

#22

Gemini-2.5-Flash

81.7%

#23

DeepSeek-R1

80.3%

#24

Gemini-2.0-Flash

78.6%

#25

GPT-4.1

78.3%

#26

Qwen-3

76.4%

#27

Qwen-3-Thinking

76.4%

#28

Llama-3.1-405B

74.4%

#29

Gemini-1.5-Pro

73.6%

#30

GPT-4-mini

64.7%

#31

GPT-3.5-Turbo

51.7%

#32

GPT-4o-0513

49.7%

#33

Grok-3-Mini-Beta

45.0%

#34

DeepSeek-V3

43.6%

#35

Grok-3-Beta

28.9%

#36

OpenAI-O1-mini

26.1%

#37

Grok-4-Thinking

25.6%

#38

Grok-4

25.6%

#39

Mistral-Large-2

18.3%

#40

Cohere-Command-R-Plus

11.4%