text-embedding-ada-002

OpenAI’s second-generation embedding model converts text into 1536-dimensional vectors for semantic search and clustering.

Text-embedding-ada-002 replaced five separate models with a single, versatile endpoint (v2) that handles diverse tasks like code search and text similarity. It supports a 8192-token context window and outputs a fixed 1536-dimension vector, making it significantly more cost-effective than its predecessors. Developers use it to power RAG pipelines, recommendation engines, and anomaly detection by calculating cosine similarity between high-dimensional numerical representations.

https://platform.openai.com/docs/guides/embeddings