text-embedding-ada-002 is OpenAI's second-generation general-purpose embedding model, designed to convert text into dense vector representations that downstream applications can use for semantic search, clustering, recommendations, and classification. It is documented in OpenAI's API model catalog under a page dedicated to the model, confirming its position as a first-party OpenAI offering. Its practical strength, as framed by Microsoft's catalog listing, is that it consolidates a single embedding endpoint for heterogeneous tasks: text search, code search, sentence similarity, and text classification are all listed as key capabilities, removing the need to maintain separate models per retrieval or matching use case.

The same model is also surfaced through Microsoft Foundry under the "Direct from Azure" portfolio, which Microsoft describes as a curated set managed end-to-end inside Azure with unified billing, governance, support, and either pay-as-you-go or reserved capacity options. That dual availability makes the model attractive for teams that want OpenAI's embedding quality but need to deploy inside Azure's enterprise infrastructure for compliance or procurement reasons. The Foundry listing reports the catalog entry as Version 2, indicating an actively maintained revision rather than a legacy artifact, while the OpenAI documentation keeps the model reachable through the standard API surface for retrieval-oriented workloads.