You can now view real-time token and latency usage in GitHub Models (GA)

The GitHub Models playground now shows your real-time token usage, including both input and output counts, as well as latency information in its responses. The analytics are accessible via the top bar, helping you more efficiently optimize prompts, evaluate model costs, and monitor response times.

Additionally, clicking on the information in the top bar now opens a modal with more details about the token and latency metrics:

Token and latency information in GitHub Models

GitHub Models makes it easy for every developer on GitHub to build AI features and products, with access to top AI models via a playground, API, and more.

To learn more about GitHub Models, check out the docs. You can also join our dedicated community discussion to discuss this update, swap tips, and share feedback.

Organization Private Registry Configuration for Java and C# CodeQL Scans

CodeQL build-mode: none scans can now access private dependencies stored in private registries (e.g. Artifactory) for Java and C# projects. This makes your scans more comprehensive, ensuring you receive all important alerts regardless of where your dependencies are stored.

Previously, build-mode: none code scans with the default setup were unable to fetch code for dependent packages stored in private registries, which could result in incomplete analysis. Now, organization administrators can configure access credentials for private registries at the organization level. This enhancement allows CodeQL scans in child repositories to retrieve all necessary dependencies, enabling comprehensive code analysis when using the code scanning default setup.

This feature is currently in public preview for GitHub Advanced Security customers.

See more

Announcing GitHub Copilot Free

GitHub Copilot Free

With the new GitHub Copilot Free plan, anyone can experiment with GitHub Copilot in Visual Studio Code and on GitHub! By simply signing in with your personal GitHub account, you will have access to 2000 Code Completions and 50 chat messages per month!

Here are a few things to try out today:

  • Accelerate your development with code completion
  • Execute edits across multiple files with Copilot Edits
  • Choose the model that works best for you, starting with Anthropic’s Claude 3.5 Sonnet
  • Access the Copilot Extensions to customize your development experience

To get started, visit our immersive experience and learn more in our announcement blog.

Stay tuned for updates as we work to continually enhance your developer experience and be sure to share feedback in Discussions.

See more
View all changes