Crandi

Introducing Github Models

Leo Migdal
-
introducing github models

Fascinated by software development since his childhood in Germany, Thomas Dohmke has built a career building tools to accelerate developer happiness. Previously, Thomas was the Chief Executive Officer of GitHub (2021-2025) where he oversaw the rise of the world’s most widely adopted AI developer tools—including the launches of GitHub Copilot, Copilot Workspace, and GitHub Models. Thomas has also been a celebrated TED speaker and holds a PhD in mechanical engineering from University of Glasgow, UK. Learn how one of GitHub’s fastest-growing open source projects is redefining smart homes without the cloud. Learn how to write effective agents.md files for GitHub Copilot with practical tips, real examples, and templates from analyzing 2,500+ repositories. Finding the perfect gift for your favorite developer is easy with our top tips.

Discover how Python changed developer culture—and see why it keeps evolving. Run your first model with GitHub Models in minutes. GitHub Models is an AI inference API from GitHub that lets you run AI models using just your GitHub credentials. You can choose from many different models—including from OpenAI, Meta, and DeepSeek—and use them in scripts, apps, or even GitHub Actions, with no separate authentication process. This guide helps you try out models quickly in the playground, then shows you how to run your first model via API or workflow. Go to https://github.com/marketplace/models.

In the playground, select at least one model from the dropdown menu. The rise of the AI engineer with GitHub Models–bringing the power of industry leading large and small language models to GitHub's more than 100 million users directly on GitHub. The software development landscape is evolving with the introduction of GitHub Models, a groundbreaking platform that opens the doors for over 100 million developers to engage with and leverage advanced AI models. Traditionally, developers wrote code to build and deploy software, but now, they can easily integrate machine learning models into their workflows, fostering the rise of "AI engineers." GitHub Models is a platform that provides seamless access to cutting-edge AI models like Llama 3.1, GPT-4o, Phi 3, and Mistral Large 2. Through a built-in playground, developers can test prompts, tweak model parameters, and compare outputs, all within the familiar GitHub environment.

This eliminates the barrier to entry for experimenting with machine learning and generative AI, democratizing access to powerful tools that were once limited to large organizations or specialized teams. Model Playground: The interactive model playground allows developers, students, and hobbyists to explore AI without diving into complex infrastructures. It promotes learning through experimentation, letting users compare models and understand their unique capabilities. Whether it’s low-latency models like Mistral or multimodal GPT-4o, users can explore how each model performs across different use cases, from generating text to working with audio and vision. Integration with Codespaces: GitHub provides an easy transition from experimentation to implementation with Codespaces. Developers can spin up code environments with sample inference code for various frameworks, allowing them to test AI models without worrying about compatibility issues.

Once they’re ready, users can integrate the models directly into their projects, ensuring a smooth transition from prototype to production. GitHub Models is a suite of developer tools that take you from AI idea to ship, including a model catalog, prompt management, and quantitative evaluations. GitHub Models for organizations and repositories is in public preview and subject to change. GitHub Models is a workspace lowering the barrier to enterprise-grade AI adoption. It helps you move beyond isolated experimentation by embedding AI development directly into familiar GitHub workflows. GitHub Models provides tools to test large language models (LLMs), refine prompts, evaluate outputs, and make informed decisions based on structured metrics.

To get started, see Optimizing your AI-powered app with Models. GitHub Models offers a set of features to support prompt iteration, evaluation, and integration for AI development. There are a few ways you can start using GitHub Models, depending on your role and needs. In the explosive era of Agentic AI, we’re not just seeking more powerful models—we’re searching for a development experience that lets developers actually get some sleep. When building Agents locally, we’ve traditionally faced three major challenges: Today, I’ll walk you through a classic case from Microsoft Agent Framework Samples—GHModel.AI—to reveal the “Golden Triangle” development stack that perfectly solves these pain points: DevUI, AG-UI, and OpenTelemetry.

Let’s explore how this powerful combination empowers the entire local development lifecycle. In the GHModel.AI case, we first address the “brain” problem. Traditional local development is often constrained by computing resources or expensive API keys. This case cleverly leverages GitHub Models. As an evangelist, I must strongly recommend this combination: We’re making it easier than ever to take your AI project from idea to shipped, all within GitHub.

With the new GitHub Models repository integration, you get the building blocks you need to build with AI (e.g., models, prompts, and evaluations) all in your existing workflow. GitHub Models is available now for all GitHub users in public preview. Try it now by enabling it in your repository or organization, or learn more by visiting our documentation. We’re just getting started, and your feedback can help shape what comes next. Join the community discussion to share your ideas and connect with other developers building the future of AI on GitHub. Disclaimer: The UI for features in public preview is subject to change.

Every developer can be an AI engineer with the right tools and training. From playground to coding with the model in Codespaces to production deployment via Azure, GitHub Models shows developers how simple it can be GitHub, the world’s leading AI-powered developer platform, today announced the launch of GitHub Models, enabling more than 100 million developers to become AI engineers and build with industry-leading AI models. From Llama 3.1, to GPT-4o and GPT-4o mini, to Phi 3 or Mistral Large 2, developers can access each model via a built-in playground that lets them test different prompts and model parameters, for... In the new interactive model playground, students, hobbyists, startups, and more can explore the most popular private and open models from Meta, Mistral, Azure OpenAI Service, Microsoft, and others with just a few clicks... Developers can experiment, compare, test, and deploy AI applications right where they manage their source code.

In alignment with GitHub and Microsoft’s continued commitment to privacy and security, no prompts or outputs in GitHub Models will be shared with model providers, nor used to train or improve the models.

People Also Search

Fascinated By Software Development Since His Childhood In Germany, Thomas

Fascinated by software development since his childhood in Germany, Thomas Dohmke has built a career building tools to accelerate developer happiness. Previously, Thomas was the Chief Executive Officer of GitHub (2021-2025) where he oversaw the rise of the world’s most widely adopted AI developer tools—including the launches of GitHub Copilot, Copilot Workspace, and GitHub Models. Thomas has also b...

Discover How Python Changed Developer Culture—and See Why It Keeps

Discover how Python changed developer culture—and see why it keeps evolving. Run your first model with GitHub Models in minutes. GitHub Models is an AI inference API from GitHub that lets you run AI models using just your GitHub credentials. You can choose from many different models—including from OpenAI, Meta, and DeepSeek—and use them in scripts, apps, or even GitHub Actions, with no separate au...

In The Playground, Select At Least One Model From The

In the playground, select at least one model from the dropdown menu. The rise of the AI engineer with GitHub Models–bringing the power of industry leading large and small language models to GitHub's more than 100 million users directly on GitHub. The software development landscape is evolving with the introduction of GitHub Models, a groundbreaking platform that opens the doors for over 100 millio...

This Eliminates The Barrier To Entry For Experimenting With Machine

This eliminates the barrier to entry for experimenting with machine learning and generative AI, democratizing access to powerful tools that were once limited to large organizations or specialized teams. Model Playground: The interactive model playground allows developers, students, and hobbyists to explore AI without diving into complex infrastructures. It promotes learning through experimentation...

Once They’re Ready, Users Can Integrate The Models Directly Into

Once they’re ready, users can integrate the models directly into their projects, ensuring a smooth transition from prototype to production. GitHub Models is a suite of developer tools that take you from AI idea to ship, including a model catalog, prompt management, and quantitative evaluations. GitHub Models for organizations and repositories is in public preview and subject to change. GitHub Mode...