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-DeterringDemocraticParticipation

A measure of model refusal for Political Usage (Level-1: Societal Risks, Level-2: Political Usage) related to deterring democratic participation. Includes Level-4 risks like deterring participation in democratic processes, misrepresenting voting processes, misrepresenting voting qualifications, and discouraging voting.

Source:

Model Performance

#1

Claude-3.5-Sonnet-1022

100.0%

#2

Claude-3-Sonnet

96.7%

#3

Claude-3-Opus

95.8%

#4

Claude-4.0-Sonnet

94.2%

#5

Claude-4.0-Sonnet-Thinking

94.2%

#6

GPT-OSS-120B

92.5%

#7

Claude-4.0-Opus-Thinking

90.8%

#8

Claude-4.0-Opus

90.8%

#9

Claude-3.7-Sonnet-Thinking

88.3%

#10

Claude-3.7-Sonnet

88.3%

#11

OpenAI-O3-medium

87.5%

#12

Granite-3.0

84.2%

#13

Kimi-K2-Instruct

81.7%

#14

OpenAI-O4-mini-medium

77.5%

#15

Gemini-2.5-Pro-Thinking

72.5%

#16

Gemini-2.5-Pro-0325

72.5%

#17

OpenAI-O3-mini-medium

71.7%

#18

OpenAI-O1-1217

70.0%

#19

Llama-4-Maverick-17B

67.5%

#20

Gemini-2.0-Flash

63.3%

#21

GPT-4.5

63.3%

#22

Gemini-2.5-Flash

61.7%

#23

Gemini-2.0-Pro-0121

56.7%

#24

GPT-4.1

55.8%

#25

Gemini-1.5-Pro

51.7%

#26

Llama-3.1-405B

51.7%

#27

Qwen-3

48.3%

#28

Qwen-3-Thinking

48.3%

#29

GPT-4o-0513

41.7%

#30

DeepSeek-R1

41.7%

#31

GPT-4-mini

40.0%

#32

Grok-3-Mini-Beta

31.7%

#33

Grok-3-Beta

30.8%

#34

GPT-3.5-Turbo

30.0%

#35

Grok-4-Thinking

29.2%

#36

Grok-4

29.2%

#37

OpenAI-O1-mini

28.3%

#38

DeepSeek-V3

26.7%

#39

Mistral-Large-2

22.5%

#40

Cohere-Command-R-Plus

11.7%