o1-preview

Approved Data Classifications

Description

The o1-preview model from OpenAI represents a significant advancement in AI reasoning capabilities, designed to tackle complex tasks with enhanced cognitive processes. By employing techniques such as Chain of Thoughts and Self Reflection, o1-preview mimics human-like reasoning, allowing it to spend more time analyzing problems before generating responses. This approach has led to remarkable achievements, including solving 83% of problems in the American Invitational Math Examination and performing at PhD-level accuracy in STEM subjects. Unlike its predecessors, o1-preview simplifies user interaction by requiring less intricate prompt engineering, making it accessible for a broader range of applications while maintaining a focus on safety and adherence to content policies.

Capabilities

Model	Training Data	Input	Output	Context Length	Cost (per 1 million tokens)
o1-preview	May 2024	Text	Text	128,000	$15.00/1M input $60.00/1M output

info

• 1M represents 1 Million Tokens
• All prices listed are based on 1 Million Tokens

Availability

Cloud Provider

Usage

curl

curl -X POST https://api.ai.it.ufl.edu/v1/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer <API_TOKEN>" \
-d '{
    "model": "o1-preview",
    "messages": [
        {
            "role": "system",
            "content": "You are a helpful assistant."
        },
        {
            "role": "user",
            "content": "Write a haiku about an Alligator."
        }
    ]
}'

python

from openai import OpenAI
  client = OpenAI(
      api_key="your_api_key",
      base_url="https://api.ai.it.ufl.edu/v1"
  )

  response = client.chat.completions.create(
      model="o1-preview", # model to send to the proxy
      messages = [
          { role: "system", content: "You are a helpful assistant." },
          {
              "role": "user",
              "content": "Write a haiku about an Alligator."
          }
      ]
  )

  print(response.choices[0].message)

javascript

import OpenAI from 'openai';

const openai = new OpenAI({
  apiKey: 'your_api_key',
  baseURL: 'https://api.ai.it.ufl.edu/v1'
});

const completion = await openai.chat.completions.create({
    model: "o1-preview",
    messages: [
        { role: "system", content: "You are a helpful assistant." },
        {
            role: "user",
            content: "Write a haiku about an Alligator.",
        },
    ],
});

print(completion.choices[0].message)

When to use

You might choose to use the o1-preview model when you require a high level of reasoning and analytical capability for complex tasks, particularly in fields such as mathematics, science, and technical problem-solving. Its advanced cognitive features make it ideal for applications that demand precise calculations, detailed explanations, or nuanced understanding of intricate concepts. Additionally, if your project involves generating content that requires a structured thought process—such as academic writing, coding assistance, or strategic planning—o1-preview can provide valuable insights and solutions. Moreover, its user-friendly interaction model allows for straightforward engagement without the need for extensive prompt engineering, making it suitable for both experienced users and those new to AI technology. Whether you're developing educational tools, conducting research, or seeking innovative solutions to challenging problems, o1-preview offers the robust performance needed to achieve your goals effectively.