GitHub Releases
GitHub changelog, platform updates, and Copilot coding-assistant releases. New features, breaking changes, security advisories and deprecations - each summarised in plain English and updated continuously.
Tracking 199 GitHub releases · Updated
- GitHub Changelog aigapreviewengineer ·
GitHub Copilot for Jira reaches General Availability
GitHub Copilot for Jira has launched as generally available, incorporating feedback from its March 2026 preview. The update introduces real-time agent progress streaming within Jira issues and post-session steering for pull requests, allowing continued instructions without creating new pull requests. Onboarding has also been simplified, reducing setup steps for connecting GitHub organizations and repositories.
feature announcement - GitHub Changelog aigapreviewengineer ·
GitHub Copilot for Jira reaches General Availability
GitHub Copilot for Jira is now generally available after a public preview that began in March 2026. This release includes real-time progress monitoring within Jira issues and post-session steering for agents to refine draft pull requests, streamlining the development workflow. Onboarding has also been simplified, requiring fewer configuration steps to connect GitHub organizations and repositories. The update is available via the Atlassian Marketplace for existing and new users.
feature announcement - GitHub Changelog aiengineer ·
Copilot Code Review: Deeper Analysis and Efficiency Gains
Copilot code review now leverages built-in CLI file exploration tools for improved cost efficiency without workflow changes. Users in the Medium analysis depth public preview gain new configurability and visibility options, including attribution in PR comments and organization-level default level settings. These enhancements reduce review costs by approximately 20% while maintaining review quality, benefiting developers and architects using Copilot for code reviews.
feature patch - GitHub Changelog aigaengineereducation ·
Copilot Free/Student Plans Default to Auto Model Selection
GitHub Copilot's Free and Student plans now default to automatic model selection, simplifying the user experience by dynamically choosing the best model for tasks. This change provides access to multiple model families while removing manual choice and the 'Preview' label for Microsoft-released models. Users can find more details in the documentation, and discussions are available on the GitHub Community.
feature patch - GitHub Changelog aiazureengineer ·
GitHub Copilot App Adds Bring Your Own Key (BYOK) Support
The GitHub Copilot app now supports BYOK, allowing users to connect their own model providers like Azure OpenAI, Anthropic, and self-hosted options. This feature enhances flexibility by enabling users to choose models based on session needs while keeping traffic within their own cloud tenants. BYOK is available for Copilot Business and Enterprise users who have the Copilot CLI enabled.
feature - GitHub Changelog aigaengineer ·
GitHub Copilot CLI terminal interface now GA
GitHub Copilot CLI's redesigned terminal interface is now generally available, offering a tabbed layout for interacting with issues, pull requests, and gists directly from the command line. The update includes a guided, in-session experience for configuring tools like MCP servers, skills, and plugins, along with a cleaner, more accessible UI with theme-aware colors and screen reader support. Users can update the CLI by running `copilot update` in their terminal.
feature patch - GitHub Changelog aipreviewengineer ·
GitHub Copilot for JetBrains IDEs adds Claude, organization agents, CLI queues
GitHub Copilot for JetBrains IDEs now offers Claude as an agent provider in public preview, alongside support for organization and enterprise custom agents. Users can queue and steer messages in Copilot CLI sessions for better control over long-running tasks. These updates aim to enhance agent flexibility, standardization, and user control for developers using these IDEs.
feature announcement - GitHub Changelog aiengineer ·
GitHub Copilot usage API now shows AI credits consumed per user
GitHub's Copilot usage metrics API now includes a new field, `ai_credits_used`, which reports the total AI credits consumed by each user daily. This enhancement allows enterprise administrators and organization owners to better connect AI consumption to value, understand adoption across teams, and plan for usage-based billing by monitoring day-over-day patterns. These metrics are available to those with REST API access for user-level reports at enterprise and organization levels.
feature - GitHub Changelog aideprecationengineer ·
GitHub Copilot deprecates Opus 4.6 (fast) model
GitHub Copilot will deprecate the Opus 4.6 (fast) model across all its experiences on June 29th, 2026. Users should migrate to the Opus 4.8 (fast) model to avoid disruption. Administrators may need to enable the new model in Copilot settings.
deprecation - GitHub Changelog aiengineer ·
MAI-Code-1-Flash expands to more GitHub Copilot features
Microsoft's MAI-Code-1-Flash coding model is now integrated into additional GitHub Copilot surfaces, including Copilot CLI, the GitHub Copilot app, and Copilot Chat on GitHub. This small, high-quality model is designed to improve coding assistance across various platforms and plans. Availability will gradually expand, with business and enterprise access planned soon.
feature announcement - GitHub Changelog aigaengineer ·
Copilot code review adds AGENTS.md support and UI improvements
GitHub Copilot code review now supports AGENTS.md files at the repository root, allowing users to shape feedback based on project conventions. UI enhancements include an easier way to request reviews on draft pull requests and collapsed timeline events for a cleaner pull request conversation view. These updates are generally available and aimed at improving the code review experience for developers.
feature patch - GitHub Changelog aipreviewengineer ·
GitHub Issues: Duplicate Detection Preview and Issue Fields MCP Support
GitHub Issues is introducing a public preview for duplicate issue detection, which flags potential matches against existing issues as they are created to reduce maintainer triage time. Additionally, issue fields now support MCP server AI tools, enabling agents to automatically set and filter issues by various fields. These features aim to streamline repository maintenance and issue management for maintainers and developers.
feature announcement - GitHub Changelog aigaengineer ·
GitHub pull request search includes Copilot-authored PRs
GitHub's pull request search now includes pull requests authored by the Copilot cloud agent when using author searches like `author:@me`. This enhancement allows users to view and manage all pull requests they are responsible for, regardless of whether they were created directly or with Copilot's assistance. The change is available in the GitHub UI and Mobile, with API support rolling out on July 16.
feature - GitHub Changelog aiengineer ·
GitHub Copilot PRs now credit users in generated release notes
GitHub's generated release notes will now credit the user who initiated a pull request created by Copilot, not just the Copilot agent itself. This change provides better recognition for developers' contributions when using Copilot for pull requests. The update is available now for all GitHub repositories and plans.
feature - GitHub Changelog aiengineereducation ·
GitHub Copilot Individual Plan Sign-ups Reopening
GitHub is gradually reopening sign-ups for its Copilot individual plans, including Student, Pro, Pro+, and Max, over the next few weeks. This announcement also details updates to usage limits, offering options to upgrade plans or pay for additional usage as needed. These changes aim to provide users with more flexibility and better manage their Copilot experience.
announcement - GitHub Changelog aigaengineer ·
GitHub Copilot Chat adds auto model selection for all users
GitHub Copilot Chat now offers auto model selection for all users on github.com and the mobile app, allowing it to dynamically choose models based on request complexity and availability. This feature aims to optimize token usage and response quality, benefiting all Copilot plan subscribers. Auto mode routes to various models including Claude Sonnet 4.6, GPT-5.4 mini, GPT-5.4, and Haiku 4.5, with usage billed at a 10% discount.
feature announcement - GitHub Changelog aiengineer ·
GitHub Copilot Agent Finder enables dynamic capability discovery
GitHub Copilot now includes an Agent Finder feature, which dynamically discovers and suggests relevant AI resources for tasks described in plain language, eliminating the need for manual configuration. This enhancement improves efficiency by loading only necessary tools and offering ranked matches, benefiting all GitHub Copilot users. The feature supports private registries and offers managed settings for enterprise control, with no automatic installations.
feature - GitHub Changelog aigaengineer ·
GitHub Copilot App Now Generally Available
The GitHub Copilot app, a desktop environment for agent-driven development, is now generally available for macOS, Windows, and Linux. New features include Canvases for shared interaction, cloud automations for scheduled work, and support for custom models and tools. Organization administrators must enable Copilot CLI for Business/Enterprise users.
feature announcement - GitHub Changelog aideprecationengineer ·
GitHub Models unavailable to new customers as retirement nears
GitHub Models is being retired, and new customers can no longer access it, even on free or paid plans. Existing users are unaffected for now, but the service will eventually be fully retired. Azure AI Foundry is recommended as an alternative for new projects needing AI model access.
deprecation announcement - GitHub Changelog aiengineer ·
GitHub Copilot usage metrics now include more active users
GitHub Copilot enterprise usage reports now incorporate server-side telemetry alongside client signals, increasing the count of active users displayed. This change addresses inconsistencies caused by client-side telemetry failures, providing a more complete picture of user engagement. While top-level counts are more accurate, detailed per-interaction breakdowns for these newly surfaced users are still limited but expected to improve in future releases.
feature
About GitHub release tracking on ReleaseBytes
The GitHub changelog covers Actions, Copilot, code security, the API and the platform itself — dozens of entries a week. ReleaseBytes summarises each changelog entry and tags breaking changes and deprecations (like Actions runner image retirements and API sunsets) so platform teams catch them before workflows break.
Frequently asked questions
How often are GitHub release notes updated on ReleaseBytes? ›
Continuously. ReleaseBytes monitors the official GitHub release channels around the clock and publishes a plain-English summary of each announcement shortly after it lands.
What kinds of GitHub changes does ReleaseBytes track? ›
New features, enhancements, bug fixes, security advisories, breaking changes, deprecations and end-of-life announcements. Every item is tagged by type so you can filter to just the changes that need action.
How can I get alerts for new GitHub releases? ›
Set up a free email or Slack alert filtered to GitHub, subscribe to the weekly digest, or follow the RSS feed. Teams can also install the ReleaseBytes GitHub App or connect via MCP.
Where does the GitHub release data come from? ›
From the official sources: GitHub changelog, platform updates, and Copilot coding-assistant releases. Every item links back to the original vendor announcement.