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-PoliticalPersuasion

A measure of model refusal for Political Usage (Level-1: Societal Risks, Level-2: Political Usage) related to political persuasion. Includes Level-4 risks like lobbying, generating high-volume campaign materials, personalized or targeted campaign materials, building systems for political campaigning or lobbying, building products for political campaigning or lobbying, political advertisements, and propaganda.

Source:

Model Performance

#1

Claude-3-Sonnet

93.3%

#2

GPT-OSS-120B

93.3%

#3

OpenAI-O1-1217

92.9%

#4

OpenAI-O3-medium

92.4%

#5

Claude-3.5-Sonnet-1022

91.0%

#6

OpenAI-O4-mini-medium

88.1%

#7

Claude-3-Opus

85.7%

#8

OpenAI-O3-mini-medium

79.0%

#9

Granite-3.0

74.3%

#10

Claude-4.0-Sonnet

67.1%

#11

Claude-4.0-Sonnet-Thinking

67.1%

#12

Kimi-K2-Instruct

61.9%

#13

GPT-4.5

61.4%

#14

Claude-4.0-Opus-Thinking

59.0%

#15

Claude-4.0-Opus

59.0%

#16

Llama-4-Maverick-17B

57.6%

#17

Claude-3.7-Sonnet-Thinking

56.2%

#18

Claude-3.7-Sonnet

56.2%

#19

Llama-3.1-405B

49.0%

#20

Gemini-2.5-Pro-Thinking

47.6%

#21

Gemini-2.5-Pro-0325

47.6%

#22

Gemini-2.0-Pro-0121

45.7%

#23

Gemini-2.0-Flash

44.3%

#24

Gemini-1.5-Pro

43.8%

#25

GPT-4.1

40.0%

#26

Gemini-2.5-Flash

36.2%

#27

GPT-4-mini

30.5%

#28

Qwen-3

22.4%

#29

Qwen-3-Thinking

22.4%

#30

DeepSeek-R1

21.4%

#31

GPT-3.5-Turbo

20.5%

#32

GPT-4o-0513

19.0%

#33

OpenAI-O1-mini

16.2%

#34

Grok-3-Beta

12.4%

#35

Grok-3-Mini-Beta

12.4%

#36

DeepSeek-V3

7.6%

#37

Cohere-Command-R-Plus

6.2%

#38

Grok-4

4.8%

#39

Grok-4-Thinking

4.8%

#40

Mistral-Large-2

4.8%