llmboard.ai
Benchmarks
CompareRankings
llmboard.ai
Benchmarks
CompareRankings
HomeModelso3

OpenAI model product

o3

OpenAI's most powerful reasoning model. o3 is a well-rounded and powerful model across domains. It sets a new standard for math, science, coding, and visual reasoning tasks. It also excels at technical writing and instruction-following. Use it to think through multi-step problems that involve analysis across text, code, and images.

Updated Aug 10, 2026. Default version: o3

Compare
LLMBoard score55.8o3
Coverage100%7 benchmark families
Context window200KTokens
Official input price$2OpenAI API

On this page

  • Specification
  • Capability
  • Benchmarks
  • Arena
  • Pricing
  • Versions
  • About
  • Compare
  • Similar models
  • FAQ

Model specification

Structured fields from the published default version.

Version
o3
Released
Apr 16, 2025
Knowledge cutoff
May 31, 2024
Parameters
N/A
Context window
200K
Max output
100K
Inputs
image, pdf, text
Outputs
text
Open weights
No
License
Proprietary

Capability profile

This profile uses the latest version under this unique model that has a calculated LLMBoard score. Arena and price are excluded.

o3 category scores

Benchmark results

Published benchmark records for the scored version o3.

22 rows
Columns

Show columns

Aider-Polyglot0.832290.5%CAug 7, 2026
COLLIE1.031077.8%CAug 7, 2026
MathVista0.933994.7%CAug 7, 2026
ARC-AGI0.94750.0%CAug 7, 2026
AIME 20240.965390.4%CAug 7, 2026
Tau-bench0.6660.0%CAug 7, 2026
MMMU0.876390.3%CAug 7, 2026
Tau2 Retail0.882672.0%CAug 7, 2026
ERQA0.692363.6%CAug 7, 2026
Tau2 Airline0.692363.6%CAug 7, 2026
Multi-Challenge0.6102967.9%CAug 7, 2026
FrontierMath0.2131725.0%CAug 7, 2026
VideoMMMU0.8132652.0%CAug 7, 2026
ARC-AGI v20.115166.7%BAug 7, 2026
CharXiv-R0.8224754.4%CAug 7, 2026
MMMU-Pro0.8286557.8%CAug 7, 2026
Tau2 Telecom0.6303514.7%CAug 7, 2026
BrowseComp0.5445824.6%CAug 7, 2026
AIME 20250.95811449.6%CAug 7, 2026
SWE-Bench Verified0.76410438.8%CAug 7, 2026
GPQA0.86623372.0%CAug 7, 2026
Humanity's Last Exam0.1719223.1%CAug 7, 2026

Arena results

Preference and agent-evaluation signals from published Arena datasets.

95 rows
Columns

Show columns

vision style controlentity recognition131233.5500N/AAug 6, 2026
visioncaptioning171209.8545N/AAug 6, 2026
visioncreative writing191197.21,880N/AJan 9, 2026
vision style controlcaptioning201181.3545N/AAug 6, 2026
vision style controlcreative writing201191.81,880N/AJan 9, 2026
visionentity recognition211236.6500N/AAug 6, 2026
text style controljapanese441426.91,281N/AAug 6, 2026
text style controlpolish461458.53,779N/AAug 6, 2026
vision style controlhumor481211.31,666N/AAug 6, 2026
visionhumor501221.11,666N/AAug 6, 2026
textjapanese511406.01,281N/AAug 6, 2026
text style controlindustry medicine and healthcare511474.93,326N/AAug 6, 2026
textpolish521439.93,779N/AAug 6, 2026
text style controlfrench551474.3763N/AAug 6, 2026
vision style controlhomework581247.21,983N/AAug 6, 2026
vision style controlcreative writing vision591205.81,587N/AAug 6, 2026
text style controlmath601446.83,721N/AAug 6, 2026
visioncreative writing vision611213.41,587N/AAug 6, 2026
vision style controloverall621216.846,095N/AAug 6, 2026
vision style controlenglish621217.722,585N/AAug 6, 2026
visionchinese631231.11,606N/AAug 6, 2026
visionhomework631233.51,983N/AAug 6, 2026
vision style controlchinese631225.01,606N/AAug 6, 2026
vision style controlocr641223.314,553N/AAug 6, 2026
visionoverall651215.346,095N/AAug 6, 2026
visionenglish661221.022,585N/AAug 6, 2026
textindustry medicine and healthcare671448.53,326N/AAug 6, 2026
visiondiagram671218.13,622N/AAug 6, 2026
visionocr671217.914,553N/AAug 6, 2026
vision style controldiagram671231.43,622N/AAug 6, 2026
text style controlkorean681393.91,141N/AAug 6, 2026
text style controlindustry mathematical691448.13,446N/AAug 6, 2026
text style controlgerman721438.51,412N/AAug 6, 2026
text style controlindustry legal and government731451.73,813N/AAug 6, 2026
textfrench781444.6763N/AAug 6, 2026
text style controlindustry life and physical and social science801453.69,990N/AAug 6, 2026
textgerman831419.51,412N/AAug 6, 2026
text style controlnon english831423.629,499N/AAug 6, 2026
text style controlrussian831430.43,799N/AAug 6, 2026
textkorean871370.31,141N/AAug 6, 2026
textmath911424.63,721N/AAug 6, 2026
text style controloverall911430.859,661N/AAug 6, 2026
text style controlexclude ties911423.843,450N/AAug 6, 2026
textindustry legal and government931428.73,813N/AAug 6, 2026
text style controlindustry business and management and financial operations931425.89,565N/AAug 6, 2026
text style controlenglish961437.830,106N/AAug 6, 2026
text style controlindustry entertainment and sports and media981394.810,576N/AAug 6, 2026
textrussian991408.23,799N/AAug 6, 2026
text style controlchinese991463.62,967N/AAug 6, 2026
textindustry mathematical1011422.03,446N/AAug 6, 2026
textnon english1031403.829,499N/AAug 6, 2026
text factualityindustry entertainment and sports and media1031385.4671N/AAug 6, 2026
text style controlexpert1031444.92,953N/AAug 6, 2026
textindustry life and physical and social science1071427.99,990N/AAug 6, 2026
text style controlhard prompts english1081445.913,941N/AAug 6, 2026
text style controlhard prompts1091440.125,721N/AAug 6, 2026
text style controlindustry software and it services1091453.819,573N/AAug 6, 2026
text style controlspanish1091404.61,069N/AAug 6, 2026
text factualitymulti turn1121404.7627N/AAug 6, 2026
text factualityindustry life and physical and social science1131423.3510N/AAug 6, 2026
text style controlmulti turn1141418.99,933N/AAug 6, 2026
text style controlindustry writing and literature and language1151396.113,281N/AAug 6, 2026
textindustry entertainment and sports and media1161373.810,576N/AAug 6, 2026
text factualityindustry writing and literature and language1161380.5891N/AAug 6, 2026
text style controlcoding1161459.411,733N/AAug 6, 2026
text style controlcreative writing1181382.67,698N/AAug 6, 2026
textoverall1191409.459,661N/AAug 6, 2026
text factualityindustry business and management and financial operations1191389.3772N/AAug 6, 2026
text style controlinstruction following1191401.915,491N/AAug 6, 2026
textexclude ties1201392.743,450N/AAug 6, 2026
textchinese1211437.32,967N/AAug 6, 2026
textmulti turn1231404.79,933N/AAug 6, 2026
textenglish1241414.430,106N/AAug 6, 2026
text factualityinstruction following1241375.61,176N/AAug 6, 2026
text factualitynon english1241395.43,336N/AAug 6, 2026
text factualityhard prompts english1261420.21,146N/AAug 6, 2026
text factualitycoding1271421.9920N/AAug 6, 2026
text factualityexclude ties1271391.25,207N/AAug 6, 2026
textindustry business and management and financial operations1281392.39,565N/AAug 6, 2026
textspanish1281384.31,069N/AAug 6, 2026
text factualityoverall1281403.77,543N/AAug 6, 2026
text factualityenglish1281409.73,274N/AAug 6, 2026
textexpert1301399.82,953N/AAug 6, 2026
text factualityindustry software and it services1301420.61,541N/AAug 6, 2026
text factualitylonger query1301377.6978N/AAug 6, 2026
textcreative writing1331359.17,698N/AAug 6, 2026
textindustry software and it services1331414.519,573N/AAug 6, 2026
text factualityhard prompts1331408.93,686N/AAug 6, 2026
textindustry writing and literature and language1351371.713,281N/AAug 6, 2026
texthard prompts1361401.925,721N/AAug 6, 2026
text style controllonger query1361408.511,433N/AAug 6, 2026
texthard prompts english1371406.413,941N/AAug 6, 2026
textcoding1381408.011,733N/AAug 6, 2026
textinstruction following1431367.615,491N/AAug 6, 2026
textlonger query1531370.511,433N/AAug 6, 2026

Pricing

Official vendor API PAYG pricing is summarized first. The table then lists individual provider offerings without treating their minimum as the official price.

Official API
$2 input, $8 output per 1M
Official provider
OpenAI
Lowest third-party
N/A
Tracked offerings
1
1 rows
Columns

Show columns

OpenAIo3global$2$8200KAug 7, 2026

Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.

o3 versions

All published versions linked to this unique model. The score columns identify the version used by the current overall ranking.

1 rows
Columns

Show columns

o3Apr 16, 202555.8N/A200K100KNoProprietary

What is o3?

A concise description based on the published model registry.

OpenAI's most powerful reasoning model. o3 is a well-rounded and powerful model across domains. It sets a new standard for math, science, coding, and visual reasoning tasks. It also excels at technical writing and instruction-following. Use it to think through multi-step problems that involve analysis across text, code, and images.

Use the benchmark, Arena and pricing sections above as separate evidence. A missing field means the current data snapshot does not support that claim.

Data snapshot: 2026-08-07. Editorial model content is not available in the backend.

o3 vs nearby models

Open a comparison with the three ranked models immediately above and below this model.

o3vsGPT-5.4-minio3vsMiniMax M1 80Ko3vsMiMo V2 Flasho3vsGLM 4.5 Airo3vsNova 2 Liteo3vso4 mini

Models similar to o3

Recommendations prioritize the same model type and family, then the closest published LLMBoard score.

#67-0.8
OP

o4 mini

OpenAI

55.0 LLMBoard

DetailsCompare
#61+0.8
OP

GPT-5.4-mini

OpenAI

56.5 LLMBoard

DetailsCompare
#60+1.1
OP

GPT-OSS-120B

OpenAI

56.9 LLMBoard

DetailsCompare
#56+3.0
OP

GPT-4.5

OpenAI

58.8 LLMBoard

DetailsCompare
#81-4.6
OP

GPT-4-Turbo

OpenAI

51.1 LLMBoard

DetailsCompare
#49+5.8
OP

o3 mini

OpenAI

61.6 LLMBoard

DetailsCompare

FAQ

Common questions about o3.

When was o3 released?

o3's default version was released on Apr 16, 2025.

How much does o3 cost?

o3's official API price is $2 per million input tokens and $8 per million output tokens via OpenAI.

Who created o3?

o3 is published under OpenAI in the model registry.

What is the context window for o3?

The default version has a 200K token context window.

Is o3 open weight?

No. The default version is not marked as having publicly available weights.

How many API providers offer o3?

1 published provider offerings are linked to the default version.

What models should I compare o3 with?

Nearby ranked alternatives include GPT-5.4-mini, MiniMax M1 80K, MiMo V2 Flash.

Rankings

OverallCodingText ArenaPricing

Modalities

Image GenerationVideo GenerationSpeech-to-TextEmbeddings

Benchmarks

All BenchmarksReasoningMathCoding

Vendors

All VendorsOpenAIAnthropicGoogle
llmboard.aiCopyright 2026 llmboard.ai