Deprecation of Azure endpoint for GitHub Models

As of July 17, 2025 usage of https://models.inference.ai.azure.com for GitHub Models inference and embeddings is officially deprecated.

This change follows the launch of the GitHub Models API on May 15, 2025, which offers a fully supported, billable, and enterprise-ready way to access GitHub-hosted models through https://models.github.ai.

If you’re currently using https://models.inference.ai.azure.com, requests will temporarily continue to function, but we will remove support on October 17, 2025. After that date, GitHub Models usage must go through https://models.github.ai to receive valid responses.

To ensure access to the latest features, authentication, and accurate billing, all users should transition to the new GitHub Models API as soon as possible.

Learn more about GitHub Models, our new GitHub Models REST API, or join the community discussion to help guide our roadmap!