o1 - In-Depth Overview

OpenAI · o-series

outdated

Model ID: o1

Get all the details on o1, an AI model from OpenAI. This page covers its token limits, pricing structure, key capabilities such as structured_outputs, reasoning, streaming, available API code samples, and performance strengths.

Key Metrics

Input Limit

200K tokens

Output Limit

100K tokens

Input Cost

$15.00/1M

Output Cost

$60.00/1M

Sample API Code

import openai

client = openai.OpenAI()

chat_completion = client.chat.completions.create(
    model="o1",
    messages=[
        {"role": "user", "content": "Hello, how are you?"}
    ]
)
print(chat_completion.choices[0].message.content)

Required Libraries

openai

Benchmarks

Benchmark	Score	Source	Notes
lmarena text	1347	OpenLLM Leaderboard	-
lmarena webdev	1045	OpenLLM Leaderboard	-
lmarena vision	1231	OpenLLM Leaderboard	-
grind	57.1%	Vellum LLM Leaderboard	-
aime 2024	79.2%	Vellum LLM Leaderboard	-
gpqa	75.7%	Vellum LLM Leaderboard	-
swe bench	48.9%	Vellum LLM Leaderboard	-
math 500	96.4%	Vellum LLM Leaderboard	-
bfcl	67.87%	Vellum LLM Leaderboard	-
alder polyglot	61.7%	Vellum LLM Leaderboard	-

Notes

The o1 series of models are trained with reinforcement learning to perform complex reasoning. o1 models think before they answer, producing a long internal chain of thought before responding to the user.