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

Chatbot Arena (Win Rate)

Measures the model's performance in head-to-head comparisons with other models in the Chatbot Arena, where human judges evaluate the quality of responses.

Source:

Model Performance

#1

Gemini-2.5-Pro-Thinking

95.3%

#2

Gemini-2.5-Pro-0325

95.3%

#3

Gemini-2.5-Pro-0605

95.3%

#4

Grok-4

90.4%

#5

Grok-4-Thinking

90.4%

#6

GPT-5

90.3%

#7

GPT-5-Thinking

90.3%

#8

Qwen-3-Max-Preview

90.1%

#9

OpenAI-O3-medium

89.1%

#10

Grok-3-Beta

89.0%

#11

Qwen-3-Thinking

88.0%

#12

Qwen-3

88.0%

#13

Gemini-2.5-Flash

87.3%

#14

Gemini-2.0-Pro-0121

85.4%

#15

Gemini-2.0-Flash-Thinking

85.3%

#16

Gemini-2.0-Flash-Thinking-0121

85.3%

#17

GPT-4.1

83.0%

#18

Kimi-K2-Instruct

83.0%

#19

DeepSeek-V3-0324

82.4%

#20

DeepSeek-V3

82.4%

#21

Claude-4.0-Opus-Thinking

82.3%

#22

GPT-5-mini-Thinking

82.1%

#23

DeepSeek-R1

81.9%

#24

Gemini-2.0-Flash

81.4%

#25

Mistral-Medium-3.1

81.3%

#26

Grok-3-Mini-Beta

80.9%

#27

Claude-4.0-Opus

80.9%

#28

OpenAI-O1-1217

80.9%

#29

OpenAI-O4-mini-high

80.3%

#30

OpenAI-O4-mini-medium

80.3%

#31

Claude-4.0-Sonnet-Thinking

78.7%

#32

Llama-3.3-70B

77.1%

#33

GPT-4-mini

76.9%

#34

OpenAI-O3-mini-high

76.9%

#35

Claude-4.0-Sonnet

76.4%

#36

GPT-5-nano-Thinking

76.1%

#37

Cohere-Command-A

75.3%

#38

Llama-3.1-405B

74.4%

#39

Gemini-1.5-Pro

74.3%

#40

OpenAI-O1-mini

74.0%

#41

OpenAI-O3-mini-medium

74.0%

#42

Claude-3.7-Sonnet-Thinking

73.7%

#43

GPT-4o-0513

71.7%

#44

Claude-3.7-Sonnet

71.6%

#45

Claude-3.5-Sonnet-1022

71.3%

#46

Claude-3-Opus

66.4%

#47

Cohere-Command-R-Plus

61.9%

#48

Claude-3-Sonnet

60.4%

#49

GPT-3.5-Turbo

49.3%

#50

Llama-2-13B

41.9%