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-MilitaryAndWarfare

A measure of model refusal for Violence & Extremism (Level-1: Content Safety Risks, Level-2: Violence & Extremism) related to military and warfare. Includes Level-4 risks related to military operations and warfare.

Source:

Model Performance

Grok 4

100.0%

Grok 4 (Thinking)

100.0%

Claude 3.5 Haiku

98.9%

Claude 3 Sonnet

96.7%

OpenAI o1

96.7%

OpenAI o4 Mini (High Effort)

96.7%

OpenAI o4 Mini (Medium Effort)

96.7%

Claude 3.5 Sonnet

95.0%

GPT-5

93.3%

#10

GPT-5 (Thinking)

93.3%

#11

Kimi K2

91.7%

#12

Claude 3 Opus

90.0%

#13

Gemini 2.0 Pro

88.3%

#14

Claude Sonnet 4.5 (Thinking)

86.7%

#15

Claude Sonnet 4.5

86.7%

#16

Claude 3.7 Sonnet (Thinking)

86.7%

#17

Claude 3.7 Sonnet

86.7%

#18

GPT OSS 120B

86.0%

#19

OpenAI o3 Mini (Medium Effort)

85.0%

#20

OpenAI o3 Mini (High Effort)

85.0%

#21

GPT-4.1

83.3%

#22

Gemini 2.5 Pro

83.3%

#23

Llama 3.1 405B

83.3%

#24

Gemini 2.5 Pro (Thinking)

83.3%

#25

Llama 4 Maverick 17B

82.2%

#26

GPT-4o Mini

80.0%

#27

Gemini 1.5 Pro

78.9%

#28

Gemini 2.0 Flash

78.3%

#29

Claude 4.0 Sonnet

78.3%

#30

Claude 4.0 Opus

78.3%

#31

Gemini 2.0 Flash Thinking (Jan 2025)

78.3%

#32

Claude 4.0 Opus (Thinking)

78.3%

#33

Gemini 2.0 Flash (Thinking)

78.3%

#34

Claude 4.0 Sonnet (Thinking)

78.3%

#35

Claude Haiku 4.5

75.3%

#36

Claude Haiku 4.5 (Thinking)

75.3%

#37

Granite 3.0

74.0%

#38

Gemini 2.5 Flash

70.0%

#39

Gemini 2.5 Flash (Thinking)

70.0%

#40

Llama 3.3 70B

56.7%

#41

OpenAI o3 (High Effort)

56.0%

#42

OpenAI o3 (Medium Effort)

56.0%

#43

Qwen 3

50.0%

#44

Qwen 3 (Thinking)

50.0%

#45

GPT-4o

46.7%

#46

GPT-4.5

43.3%

#47

GPT-3.5 Turbo

40.0%

#48

OpenAI o1 Mini

36.7%

#49

Gemini 3.1 Pro Preview

28.7%

#50

Grok 3 (Thinking)

24.0%

#51

Grok 3 Mini

24.0%

#52

Grok 3

24.0%

#53

DeepSeek R1

15.0%

#54

Cohere Command A

13.3%

#55

Cohere Command R+

10.0%

#56

DeepSeek V3

10.0%

#57

Mistral Large 2

10.0%