OpenAI logo
Embedding

Text Embedding Ada 002

OpenAI Ada 002 January 1, 2022

A text embedding model that converts text into vector representations for various NLP tasks, producing 1536-dimensional embeddings.

Model Capabilities

Input

Accepted Formats

Text

Input Length

Context Window: 8K Tokens

Output

Generated Formats

Embedding

Output Length

Max Output: 1,536 Tokens

Core Features & Strengths

No specific core features listed.

What is it good for?

Primary Use Cases

Content Creation: Writing, editing, summarization, translation
Code Assistance: Programming help, debugging, code review

Best For

Business Applications: Customer service, content generation

Specifications & Pricing

Specifications

Model Size: 1B params
Architecture: Transformer
Open Source:No

Pricing

Input / 1M tk

$0.1000

Output / 1M tk

N/A

Est. cost in USD

💡 What does this cost in practice?

📖 300-page book:~200,000 tokens ≈ $0.0200
📄 Research paper:~15,000 tokens ≈ $0.0015
💬 Chat conversation:~1,000 tokens ≈ $0.0001
Priced per 1M input tokens; no output token pricing

Usage Guidance

How to Prompt This Model

✨ General Tips

🎯Be Specific: Detail the desired outcome, format, tone, and constraints.
🎭Set Context: Explain your role or the purpose (e.g., "Act as a helpful architect...").
📝Specify Format: Request bullet points, JSON, markdown, code blocks, etc.
🔄Iterate: Refine prompts based on responses. Don't expect perfection first try.

🚀 Best Practices for Text Embedding Ada 002

📋 Example Prompts

Content Creation Example

Write a professional email to a client explaining a project delay. The project is a website redesign, the delay is due to additional security requirements, and the new timeline is 2 weeks. Keep the tone apologetic but confident.