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-WeaponUsageDevelopment

A measure of model refusal for Violence & Extremism (Level-1: Content Safety Risks, Level-2: Violence & Extremism) related to weapon usage and development. Includes Level-4 risks like guns, explosives/dangerous materials, bioweapons/viruses/gain-of-function, nuclear weapons, chemical weapons, and radiological weapons.

Source:

Model Performance

Claude Haiku 4.5

100.0%

Claude Haiku 4.5 (Thinking)

100.0%

GPT-4.5

100.0%

OpenAI o3 (Medium Effort)

100.0%

OpenAI o3 (High Effort)

100.0%

OpenAI o4 Mini (High Effort)

100.0%

Gemini 3.1 Pro Preview

100.0%

Gemini 2.0 Pro

100.0%

GPT OSS 120B

100.0%

#10

OpenAI o4 Mini (Medium Effort)

100.0%

#11

Claude 3.5 Sonnet

97.2%

#12

Granite 3.0

96.7%

#13

Claude 3 Opus

96.7%

#14

OpenAI o3 Mini (High Effort)

90.0%

#15

OpenAI o3 Mini (Medium Effort)

90.0%

#16

Kimi K2

84.4%

#17

Claude Sonnet 4.5

83.9%

#18

Claude Sonnet 4.5 (Thinking)

83.9%

#19

GPT-4o Mini

81.1%

#20

Gemini 2.0 Flash Thinking (Jan 2025)

76.7%

#21

Gemini 2.0 Flash (Thinking)

76.7%

#22

Gemini 2.0 Flash

76.7%

#23

GPT-4.1

75.6%

#24

Claude 3.7 Sonnet

75.0%

#25

Claude 3.7 Sonnet (Thinking)

75.0%

#26

Claude 4.0 Sonnet (Thinking)

72.7%

#27

Claude 4.0 Sonnet

72.7%

#28

Claude 3.5 Haiku

70.0%

#29

Claude 3 Sonnet

68.0%

#30

GPT-5

66.7%

#31

GPT-5 (Thinking)

66.7%

#32

GPT-3.5 Turbo

65.0%

#33

Llama 3.1 405B

62.2%

#34

GPT-4o

60.0%

#35

Grok 3 Mini

60.0%

#36

Claude 4.0 Opus

59.3%

#37

Claude 4.0 Opus (Thinking)

59.3%

#38

Grok 4 (Thinking)

50.0%

#39

Grok 4

50.0%

#40

Llama 4 Maverick 17B

49.3%

#41

Qwen 3

48.9%

#42

Qwen 3 (Thinking)

48.9%

#43

Grok 3

48.3%

#44

Grok 3 (Thinking)

48.3%

#45

OpenAI o1

36.0%

#46

DeepSeek R1

34.4%

#47

Llama 3.3 70B

32.2%

#48

DeepSeek V3

32.2%

#49

Mistral Large 2

31.1%

#50

OpenAI o1 Mini

28.0%

#51

Gemini 2.5 Pro

25.3%

#52

Gemini 2.5 Pro (Thinking)

25.3%

#53

Gemini 2.5 Flash

24.7%

#54

Gemini 2.5 Flash (Thinking)

24.7%

#55

Gemini 1.5 Pro

22.7%

#56

Cohere Command A

21.1%

#57

Cohere Command R+

11.1%