Benchmark Explorer

Explore how models perform on various benchmarks

Benchmarks

🔍

Capability Benchmarks

AA-LCR

AIME

ARC-AGI

CaseLaw

Chatbot Arena (Win Rate)

Chatbot Arena AAII

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

MortgageTax

SAGE

SWE-bench

SciCode

SimpleBench

TaxEval

Terminal-Bench Hard

Vals Index

Vals Multimodal Index

Vibe Code Bench

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

Speed & Latency Metrics

Median Tokens/s

Cost & Pricing Metrics

Blended Price (USD/1M Tokens)

AIR-Bench-NonconsensualNudity

A measure of model refusal for Sexual Content (Level-1: Content Safety Risks, Level-2: Sexual Content) related to non-consensual nudity. Includes Level-4 risks like non-consensual intimate image (NCII) creation or distribution.

Source:

Model Performance

Claude 3.5 Sonnet

100.0%

Claude 4.0 Opus (Thinking)

100.0%

Claude 4.0 Opus

100.0%

Gemini 3.1 Pro Preview

100.0%

OpenAI o4 Mini (High Effort)

100.0%

Grok 4 (Thinking)

100.0%

OpenAI o4 Mini (Medium Effort)

100.0%

Grok 4

100.0%

Claude 3.7 Sonnet (Thinking)

93.3%

#10

GPT-4.1

93.3%

#11

Claude 3.7 Sonnet

93.3%

#12

Kimi K2

93.3%

#13

Claude 3.5 Haiku

92.9%

#14

Claude 4.0 Sonnet (Thinking)

88.0%

#15

Claude 4.0 Sonnet

88.0%

#16

Claude Haiku 4.5 (Thinking)

87.7%

#17

Claude Haiku 4.5

87.7%

#18

Gemini 2.0 Flash

86.7%

#19

Gemini 2.0 Flash (Thinking)

86.7%

#20

GPT-4o Mini

86.7%

#21

DeepSeek R1

86.7%

#22

Gemini 2.0 Flash Thinking (Jan 2025)

86.7%

#23

Granite 3.0

83.3%

#24

GPT OSS 120B

80.4%

#25

Claude 3 Sonnet

78.3%

#26

GPT-4.5

77.0%

#27

Gemini 2.0 Pro

76.2%

#28

GPT-5

68.8%

#29

GPT-5 (Thinking)

68.8%

#30

GPT-3.5 Turbo

67.9%

#31

Llama 3.1 405B

66.7%

#32

GPT-4o

66.7%

#33

Claude Sonnet 4.5 (Thinking)

63.3%

#34

Claude Sonnet 4.5

63.3%

#35

Claude 3 Opus

62.3%

#36

Llama 3.3 70B

60.0%

#37

OpenAI o3 (High Effort)

58.7%

#38

OpenAI o3 (Medium Effort)

58.7%

#39

DeepSeek V3

53.3%

#40

Llama 4 Maverick 17B

48.3%

#41

Mistral Large 2

46.7%

#42

Grok 3

46.4%

#43

Gemini 2.5 Flash

46.4%

#44

Gemini 2.5 Flash (Thinking)

46.4%

#45

Grok 3 (Thinking)

46.4%

#46

Qwen 3

45.7%

#47

Qwen 3 (Thinking)

45.7%

#48

Grok 3 Mini

40.6%

#49

Cohere Command A

40.0%

#50

OpenAI o1 Mini

30.0%

#51

Cohere Command R+

26.7%

#52

Gemini 2.5 Pro

20.0%

#53

Gemini 2.5 Pro (Thinking)

20.0%

#54

OpenAI o3 Mini (High Effort)

20.0%

#55

OpenAI o3 Mini (Medium Effort)

20.0%

#56

Gemini 1.5 Pro

18.3%

#57

OpenAI o1

15.0%