OpenAI Embedding Models is a series of new embedding models, including two new embedding models, the updated GPT-4 Turbo preview model, GPT-3.5 Turbo model, and text content moderation model. By default, data sent to the OpenAI API is not used to train or improve OpenAI models. The new embedding models offer lower pricing, including the smaller, more efficient text-embedding-3-small model and the larger, more powerful text-embedding-3-large model. Embeddings are a set of numbers that represent concepts within content, such as natural language or code. Embeddings make it easier for machine learning models and other algorithms to understand the relationships between content and perform tasks like clustering or retrieval. They support knowledge retrieval in the ChatGPT and Assistants API and many retrieval-augmented generation (RAG) development tools. text-embedding-3-small is a new, efficient embedding model. Its performance has significantly improved compared to its predecessor, text-embedding-ada-002, with an increase in the average MIRACL score from 31.4% to 44.0% and an increase in the average score on English tasks (MTEB) from 61.0% to 62.3%. The pricing of text-embedding-3-small is also 5 times lower than the previous text-embedding-ada-002 model, dropping from $0.0001 per thousand tokens to $0.00002. text-embedding-3-large is a new generation of larger embedding models that can create embeddings up to 3072 dimensions. It boasts improved performance, with an increase in the average MIRACL score from 31.4% to 54.9% and an increase in the average MTEB score from 61.0% to 64.6%. The pricing for text-embedding-3-large is $0.00013/thousand tokens. Additionally, we support native functionality for shortening embeddings, allowing developers to balance performance and cost.