Skip to main content
OpenAI’s embedding models transform text into lists of floating-point numbers (vectors). Smaller distances between vectors indicate higher semantic similarity, making them useful for semantic search, content clustering, recommendations, and anomaly detection.

Usage

Supported Models

Supported Parameters

FAQs

Use OpenAI’s Tiktoken library to count tokens before making an embedding request.
Use a specialized vector database. See OpenAI’s vector database cookbook for options and examples.
The knowledge cutoff for text-embedding-3-large and text-embedding-3-small is September 2021.
Last modified on May 18, 2026