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-AdultContent

A measure of model refusal for Sexual Content (Level-1: Content Safety Risks, Level-2: Sexual Content) related to adult content. Includes Level-4 risks like obscenity, suggestive content, sexual acts, and sexual intercourse.

Source:

Model Performance

#1

Claude-3.5-Sonnet-1022

96.4%

#2

Claude-4.0-Sonnet

86.2%

#3

Claude-4.0-Sonnet-Thinking

86.2%

#4

Claude-3.7-Sonnet-Thinking

85.5%

#5

Claude-3.7-Sonnet

85.5%

#6

Claude-4.0-Opus

84.1%

#7

Claude-4.0-Opus-Thinking

84.1%

#8

Granite-3.0

83.3%

#9

Gemini-2.0-Flash

81.9%

#10

GPT-OSS-120B

80.4%

#11

Claude-3-Sonnet

78.3%

#12

Gemini-2.0-Pro-0121

76.1%

#13

Llama-4-Maverick-17B

75.4%

#14

GPT-4.5

72.5%

#15

Mistral-Large-2

71.7%

#16

GPT-4o-0513

68.8%

#17

GPT-4.1

67.4%

#18

OpenAI-O1-mini

63.8%

#19

GPT-3.5-Turbo

63.0%

#20

DeepSeek-V3

63.0%

#21

GPT-4-mini

63.0%

#22

Claude-3-Opus

62.3%

#23

OpenAI-O3-medium

58.7%

#24

Llama-3.1-405B

57.2%

#25

DeepSeek-R1

55.1%

#26

Gemini-1.5-Pro

52.2%

#27

OpenAI-O4-mini-medium

51.4%

#28

Cohere-Command-R-Plus

48.6%

#29

OpenAI-O1-1217

48.6%

#30

Gemini-2.5-Pro-Thinking

46.4%

#31

Gemini-2.5-Flash

46.4%

#32

Gemini-2.5-Pro-0325

46.4%

#33

Grok-3-Beta

46.4%

#34

OpenAI-O3-mini-medium

46.4%

#35

Qwen-3

45.7%

#36

Qwen-3-Thinking

45.7%

#37

Kimi-K2-Instruct

44.2%

#38

Grok-3-Mini-Beta

40.6%

#39

Grok-4-Thinking

27.5%

#40

Grok-4

27.5%