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-Fraud

A measure of model refusal for Deception (Level-1: Societal Risks, Level-2: Deception) related to fraud. Includes Level-4 risks like spam, scams, phishing/catfishing, pseudo-pharmaceuticals, and impersonating others.

Source:

Model Performance

#1

Claude-3.5-Sonnet-1022

100.0%

#2

GPT-OSS-120B

93.3%

#3

Claude-4.0-Opus-Thinking

92.7%

#4

Claude-4.0-Opus

92.7%

#5

Claude-4.0-Sonnet-Thinking

92.0%

#6

Claude-4.0-Sonnet

92.0%

#7

OpenAI-O3-medium

91.3%

#8

Claude-3-Sonnet

88.0%

#9

Kimi-K2-Instruct

82.7%

#10

Granite-3.0

81.3%

#11

Claude-3-Opus

80.0%

#12

OpenAI-O1-1217

75.3%

#13

OpenAI-O4-mini-medium

74.7%

#14

Claude-3.7-Sonnet

72.0%

#15

Claude-3.7-Sonnet-Thinking

72.0%

#16

GPT-4.5

72.0%

#17

Llama-4-Maverick-17B

71.3%

#18

OpenAI-O3-mini-medium

66.7%

#19

GPT-4.1

62.7%

#20

Gemini-2.5-Pro-Thinking

58.7%

#21

Gemini-2.5-Pro-0325

58.7%

#22

Gemini-2.0-Pro-0121

56.0%

#23

Gemini-2.0-Flash

51.3%

#24

Gemini-2.5-Flash

43.3%

#25

Llama-3.1-405B

42.7%

#26

Gemini-1.5-Pro

41.3%

#27

GPT-4o-0513

41.3%

#28

DeepSeek-R1

38.7%

#29

Qwen-3

37.3%

#30

Qwen-3-Thinking

37.3%

#31

GPT-4-mini

37.3%

#32

Grok-3-Mini-Beta

34.7%

#33

OpenAI-O1-mini

33.3%

#34

GPT-3.5-Turbo

30.7%

#35

DeepSeek-V3

25.3%

#36

Grok-4

24.0%

#37

Grok-4-Thinking

24.0%

#38

Grok-3-Beta

21.3%

#39

Mistral-Large-2

17.3%

#40

Cohere-Command-R-Plus

15.3%