Rest Api Endpoints For Models Catalog Github Docs

Use the REST API to get a list of models available for use, including details like ID, supported input/output modalities, and rate limits. You can use the REST API to explore available models in the GitHub Models catalog. Get a list of models available for use, including details like supported input/output modalities, publisher, and rate limits. Use the REST API to submit a chat completion request to a specified model, with or without organizational attribution. You can use the REST API to run inference requests using the GitHub Models platform. The API requires the models: read scope when using a fine-grained personal access token or when authenticating using a GitHub App.

This endpoint allows you to run an inference request attributed to a specific organization. You must be a member of the organization and have enabled models to use this endpoint. The token used to authenticate must have the models: read permission if using a fine-grained PAT or GitHub App minted token. The request body should contain the model ID and the messages for the chat completion request. The response will include either a non-streaming or streaming response based on the request parameters. Setting to application/json is required.

Setting to application/vnd.github+json is recommended. Create integrations, retrieve data, and automate your workflows with the GitHub REST API. Get oriented to the REST API documentation. You can authenticate to the REST API to access more endpoints and have a higher rate limit. Follow these best practices when using GitHub's API. Learn about REST API rate limits, how to avoid exceeding them, and what to do if you do exceed them.

Use the REST API to create, manage and control the workflow of public and private GitHub repositories. You can now use the GitHub Models REST API to programmatically explore and run inference with models hosted on GitHub. This includes: These endpoints support streaming and non-streaming completions, as well as advanced options like temperature, stop sequences, and deterministic sampling via seed. Check out the Models API reference docs to get started, or join the conversation in community discussions. A Python application to interact with the GitHub Models API, demonstrating how to list available models.

GitHub Models is a recent service (announced in May 2025) that allows developers to access and experiment with AI models. The official GitHub Models API endpoints use the domain models.github.ai. According to the official documentation and our working implementation, the GitHub Models API uses these endpoints: This implementation is based on the following official GitHub documentation and announcements: Set up a virtual environment (recommended): Get oriented to the REST API documentation.

You can use GitHub's API to build scripts and applications that automate processes, integrate with GitHub, and extend GitHub. For example, you could use the API to triage issues, build an analytics dashboard, or manage releases. Each REST API endpoint is documented individually, and the endpoints are categorized by the resource that they primarily affect. For example, you can find endpoints relating to issues in REST API endpoints for issues. If you are new to REST APIs, you may find it helpful to refer to the Quickstart or Getting Started guide for an introduction. For more information, see:

If you are familiar with REST APIs but new to GitHub's REST API, you may find it helpful to refer to the authentication documentation. For more information, see: There was an error while loading. Please reload this page. There was an error while loading. Please reload this page.

I noticed that for some models, such as gpt-4.1 and gpt-4o, they use different endpoints: "https://models.github.ai/inference/chat/completions" and "https://models.inference.ai.azure.com/chat/completions" respectively, which is quite confusing. This subtle difference is easy to overlook and also increases the difficulty for developers integrating GitHub Models. Can someone explain this issue? Is it possible to provide a unified endpoint? There was an error while loading. Please reload this page.

People Also Search

• REST API endpoints for models catalog - GitHub Docs
• Models - GitHub Docs
• REST API endpoints for models inference - GitHub Docs
• GitHub REST API documentation - GitHub Docs
• REST API endpoints for repositories - GitHub Docs
• GitHub Models API now available
• GitHub Models API Examples
• About the REST API - GitHub Docs
• Why does GitHub Models have two different endpoints?
• docs/content/rest/about-the-rest-api/about-the-rest-api.md at ... - GitHub