o3 - In-Depth Overview

OpenAI · OpenAI

Current

Flagship

Latest in family

Model ID: o3

Get all the details on o3, an AI model from OpenAI. This page covers its token limits, pricing structure, key capabilities such as multimodal_input, long_context, function_calling, available API code samples, and performance strengths.

Key Metrics

Input Limit

200K tokens

Output Limit

100K tokens

Input Cost

$10.00/1M

Output Cost

$40.00/1M

Sample API Code

from openai import OpenAI
client = OpenAI()
response = client.chat.completions.create(
    model="o3",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Hello, how are you?"}
    ]
)
print(response.choices[0].message.content)

Required Libraries

openai

Benchmarks

Benchmark	Score	Source	Notes
lmarena text	1409	OpenLLM Leaderboard	Score for o3-2025-04-16
lmarena webdev	1190	OpenLLM Leaderboard	Score for o3-2025-04-16
lmarena vision	1303	OpenLLM Leaderboard	Score for o3-2025-04-16
livebench global average	80.71	LiveBench	Score for o3 High
livebench reasoning average	93.33	LiveBench	Score for o3 High
livebench coding average	76.71	LiveBench	Score for o3 High
livebench mathematics average	85.00	LiveBench	Score for o3 High
livebench data analysis average	67.02	LiveBench	Score for o3 High
livebench language average	76.00	LiveBench	Score for o3 High
livebench if average	86.17	LiveBench	Score for o3 High
aime 2024	91.6%	Vellum	-
gpqa	83.3%	Vellum	-
swe bench	69.1%	Vellum	-
humanitys last exam	20.32	Vellum	-

Notes

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.