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-PerpetuatingHarmfulStereotypes

A measure of model refusal for Hate/Toxicity (Level-1: Content Safety Risks, Level-2: Hate/Toxicity) related to perpetuating harmful beliefs. Includes Level-4 risks like negative stereotyping of any group, perpetuating racism, and perpetuating sexism.

Source:

Model Performance

#1

Claude-3.5-Sonnet-1022

95.2%

#2

Claude-4.0-Sonnet

92.1%

#3

Claude-4.0-Sonnet-Thinking

92.1%

#4

Claude-4.0-Opus

85.7%

#5

Claude-4.0-Opus-Thinking

85.7%

#6

Claude-3.7-Sonnet

80.2%

#7

Claude-3.7-Sonnet-Thinking

80.2%

#8

Claude-3-Sonnet

79.4%

#9

GPT-OSS-120B

79.4%

#10

Claude-3-Opus

77.8%

#11

Kimi-K2-Instruct

77.8%

#12

GPT-4.5

77.0%

#13

Granite-3.0

76.2%

#14

OpenAI-O3-medium

73.0%

#15

OpenAI-O4-mini-medium

69.8%

#16

GPT-4.1

69.0%

#17

Gemini-2.5-Pro-Thinking

69.0%

#18

Gemini-2.5-Pro-0325

69.0%

#19

Gemini-2.5-Flash

68.3%

#20

OpenAI-O1-1217

67.5%

#21

Gemini-2.0-Flash

65.1%

#22

DeepSeek-R1

63.5%

#23

OpenAI-O3-mini-medium

62.7%

#24

Llama-4-Maverick-17B

60.3%

#25

Gemini-2.0-Pro-0121

59.5%

#26

Gemini-1.5-Pro

54.0%

#27

Qwen-3-Thinking

54.0%

#28

Qwen-3

54.0%

#29

Llama-3.1-405B

53.2%

#30

GPT-3.5-Turbo

52.4%

#31

Grok-3-Mini-Beta

47.6%

#32

GPT-4o-0513

47.6%

#33

DeepSeek-V3

46.0%

#34

OpenAI-O1-mini

45.2%

#35

GPT-4-mini

44.4%

#36

Grok-3-Beta

42.9%

#37

Grok-4-Thinking

42.1%

#38

Grok-4

42.1%

#39

Mistral-Large-2

31.0%

#40

Cohere-Command-R-Plus

22.2%