GitHub Copilot Changelog

GitHub has retired the Copilot Billing Preview app, transitioning users to manage their Copilot expenses directly in GitHub billing settings. The new interface offers enhanced features such as user-level budgets and detailed usage reports, improving visibility over AI usage and spending.

GitHub's retirement of the Copilot Billing Preview app and the introduction of user-level budgets and detailed usage reports in its billing settings provide builders and PMs with better control over AI-related expenses. This shift enhances financial oversight, allowing teams to allocate resources more effectively and make informed decisions about AI tool investments.

GitHub Spark will cease new user registrations and app creation on August 4, 2026, with existing users able to export apps until August 31, 2026. The retirement of GitHub Models on July 30, 2026, will affect apps using the llm() function, requiring users to switch to their own inference providers to maintain AI capabilities.

The upcoming deprecation of GitHub Spark and its associated models will require builders and PMs to transition their AI applications to alternative inference providers, which could disrupt existing workflows and necessitate additional development resources. Investors should be aware of the potential impact on startups relying on GitHub's infrastructure for AI capabilities, as this shift may affect their scalability and operational costs.

GitHub Copilot will deprecate several models, including Gemini 3.1 Pro and Claude Opus 4.5, on September 1, 2026. Users are advised to transition to alternatives like Gemini 3.6 Flash and Claude Opus 4.7 or 5 before this date.

The deprecation of models like Gemini 3.1 Pro and Claude Opus 4.5 in GitHub Copilot signals a critical need for builders and PMs to plan for transitions to newer models such as Gemini 3.6 Flash and Claude Opus 4.7 or 5. This shift could impact development timelines and resource allocation, making it essential for investors to understand the evolving landscape of AI tools.

GitHub has deprecated the Gemini 2.5 Pro and Gemini 3 Flash models as of July 31, 2026. Users are encouraged to transition to Gemini 3.1 Pro (Preview) and Gemini 3.6 Flash, respectively, and administrators must enable access to these alternatives in Copilot settings.

The deprecation of Gemini 2.5 Pro and Gemini 3 Flash signals a shift towards newer models, specifically Gemini 3.1 Pro and 3.6 Flash, which may offer improved features and performance. Builders and PMs should prioritize transitioning to these alternatives to ensure compatibility and leverage advancements, while investors should consider the implications for GitHub's ongoing innovation strategy.

GitHub has introduced user-based model policy targeting for Enterprise customers using Copilot Business or Enterprise licenses, allowing admins to customize AI model access for specific teams. This public preview enables finer control over model availability based on job roles, enhancing team-level governance and user experience.

GitHub's introduction of user-based model policy targeting for Enterprise customers allows for customized AI model access based on job roles, which enhances governance and user experience. This development is crucial for builders and PMs as it enables more tailored AI solutions, while investors should note its potential to improve customer retention and satisfaction in enterprise environments.

The July 2026 update for GitHub Copilot in Visual Studio introduces a new agent in public preview, built-in .NET and Azure skills, and organization-level custom instructions, enhancing user workflows and collaboration. Key features include improved Copilot Chat responses and code review capabilities, available across all plans.

The introduction of organization-level custom instructions in GitHub Copilot enhances collaboration and workflow efficiency for teams using Visual Studio, allowing for tailored coding assistance. This development signals a shift towards more integrated AI tools in software development, which can lead to increased productivity and reduced onboarding time for new developers.

The July 2026 updates for GitHub Copilot in Visual Studio Code (v1.127-v1.131) enhance agent management, multi-chat capabilities, and introduce new accessibility features. Key improvements include a redesigned Agents window, faster code review processes, and built-in dictation support, making collaboration and navigation more efficient for developers.

The July 2026 updates to GitHub Copilot in Visual Studio Code introduce enhanced agent management and multi-chat capabilities, which streamline collaboration and code review processes for developers. This development signals a shift towards more efficient workflows and improved accessibility, making it crucial for builders and PMs to adapt their tools and strategies accordingly.

Grok 4.5, xAI's latest reasoning model, is now available in GitHub Copilot, featuring a context window of 500,000 tokens and supporting text and image inputs. It excels in terminal-based coding tasks, particularly in Visual Studio Code and Copilot CLI, and is accessible to various Copilot plans, though administrators must enable it in settings.

The release of Grok 4.5 in GitHub Copilot, with a 500,000-token context window and support for text and image inputs, enhances coding efficiency, particularly for terminal-based tasks. Builders and PMs can leverage this advanced reasoning model to streamline development processes, while investors should note its potential to improve productivity in software engineering.

The latest GitHub Copilot for JetBrains update enhances workflow control with OpenTelemetry configuration, improved model management, and support for MCP servers in Claude agent flows, facilitating better observability and cost control for enterprise users.

The enhanced OpenTelemetry configuration and model management in GitHub Copilot for JetBrains allows builders and PMs to achieve better observability and cost control in enterprise applications. This development signals a shift towards more efficient AI integration in development workflows, which can attract investor interest in tools that improve productivity and operational efficiency.

Claude Opus 5, Anthropic's latest model, is now integrated into GitHub Copilot, enhancing coding tasks with improved reasoning and tool coordination. Available for Pro+, Max, Business, and Enterprise users, it features safeguards against high-harm content and is billed under usage-based pricing.

The integration of Claude Opus 5 into GitHub Copilot enhances coding efficiency through improved reasoning and tool coordination, which can significantly reduce development time for builders and PMs. For investors, this signals a growing trend in AI-assisted development tools that prioritize safety and usability, indicating potential market growth in AI-driven software solutions.

GitHub's Copilot cloud agent for Linear is now generally available, allowing users to assign issues that the agent can autonomously analyze and create draft pull requests. Users can customize the agent's model, branch settings, and provide real-time instructions directly from their Linear workspace.

The general availability of GitHub's Copilot cloud agent for Linear enables teams to automate issue analysis and draft pull requests, streamlining development workflows. This development signals a significant shift towards increased efficiency in project management and coding, allowing builders and PMs to focus on higher-level tasks while investors can anticipate improved productivity metrics.

GitHub MCP Server has adopted the upcoming stateless MCP protocol, enhancing scalability and performance by removing sessions and enabling parallel handshakes. This update supports innovative extensions and maintains backward compatibility with tier 1 SDKs, ensuring seamless integration for developers ahead of the official release in 2026.

The adoption of the stateless MCP protocol by GitHub MCP Server enhances scalability and performance, allowing builders and PMs to develop more efficient applications with parallel handshakes. For investors, this update signals a commitment to innovation and future-proofing, as it supports new extensions while ensuring compatibility with existing SDKs, potentially increasing the value of related projects.

GitHub Issues introduces agent automation controls in public preview, allowing users to approve or decline automated changes with confidence ratings and rationale. This feature enhances issue management by providing transparency and control over automation, making it easier to handle issues in busy repositories.

The introduction of agent automation controls in GitHub Issues allows users to approve or decline automated changes based on confidence ratings and rationale, enhancing issue management in busy repositories. For builders and PMs, this means improved oversight and control over automation processes, potentially increasing productivity and reducing errors in project management.

GitHub has launched a new Copilot metrics impact dashboard for enterprise admins, providing insights into user engagement across AI adoption phases. It includes metrics like average pull requests merged and offers actionable steps to enhance Copilot usage among developers.

GitHub's launch of a Copilot metrics impact dashboard allows enterprise admins to track user engagement and optimize AI tool adoption. This development provides actionable insights that can help builders and PMs enhance productivity and workflow efficiency, while investors can assess the effectiveness of AI integration in development processes.

Gemini 3.6 Flash, Google's latest model for GitHub Copilot, enhances web and app development with improved task-completion rates and token efficiency compared to Gemini 3.5 Flash. Available to various Copilot users, it supports parallel tool use and requires admin activation for enterprise plans.

The release of Gemini 3.6 Flash in GitHub Copilot improves task-completion rates and token efficiency, which can significantly enhance productivity for developers. This advancement allows teams to leverage parallel tool use, making it easier to integrate AI into their workflows, ultimately leading to faster development cycles and reduced costs.

GitHub now allows direct management of AI credit pools for cost centers in the billing UI, enhancing usability for Copilot Business and Enterprise users. This feature automatically adjusts credit limits based on assigned licenses and provides options for handling usage limits, ensuring cost centers stay within budgeted AI credits.

GitHub's introduction of AI credit pools management in the billing UI allows builders and PMs to better control AI usage costs, ensuring that teams can optimize their budgets while utilizing Copilot effectively. For investors, this development signals a commitment to enhancing user experience and operational efficiency, potentially leading to increased adoption and revenue growth in AI-driven tools.

GitHub Copilot Business and Enterprise users can now track AI credit usage per billing cycle on their usage page, regardless of budget settings. This update allows users to see actual credit consumption, enhancing transparency and management of token usage.

The introduction of AI credit tracking for GitHub Copilot Business and Enterprise users enhances transparency in resource consumption, allowing builders and PMs to better manage costs and optimize workflows. For investors, this development signals a commitment to improving user experience and operational efficiency, which could lead to increased adoption and revenue growth.

GitHub Copilot code review introduces enhanced customization with head branch instruction reading, expanded file support, and firewall functionality, giving developers more control over their review processes. Administrators can now configure environments using a copilot-code-review.yml file, while firewall settings ensure secure reviews by default.

The introduction of enhanced customization in GitHub Copilot's code review, including the ability to configure environments with a copilot-code-review.yml file and improved firewall settings, allows builders and PMs to tailor the review process to their specific needs, enhancing productivity and security. For investors, this signals a commitment to continuous improvement in developer tools, potentially increasing user adoption and retention.

The June 2026 update for GitHub Copilot in Visual Studio enhances visibility and trust, introducing usage tracking, a trust validation layer for MCP servers, and general availability of the C++ modernization agent. Users can now also add pull requests to Copilot Chat and review them directly within the IDE.

The June 2026 update for GitHub Copilot in Visual Studio introduces usage tracking and a trust validation layer, which enhances developer accountability and security in collaborative coding environments. This is significant for builders and PMs as it improves workflow efficiency and trust in AI-assisted development, while investors should note the potential for increased adoption of AI tools in software engineering.

GitHub Copilot for JetBrains now supports custom endpoints for BYOK, enhanced plugin management, and Claude agent customizations. The update also introduces local sandboxing, built-in debugging for Copilot CLI, and various user experience improvements, aiming to provide teams with greater flexibility in software development.

The expansion of GitHub Copilot for JetBrains to support custom endpoints for BYOK (Bring Your Own Key) enhances security and flexibility for development teams, allowing them to integrate AI tools while maintaining control over their data. This development is significant for builders and PMs as it streamlines workflows and improves productivity, while investors should note the growing emphasis on secure AI integrations in software development.

GitHub Copilot now includes a /security-review command in public preview, allowing users to analyze code changes for vulnerabilities directly within the app. This feature provides high-confidence findings, actionable suggestions, and prioritizes issues, making it easier to address security concerns before code deployment.

The introduction of the /security-review command in GitHub Copilot allows builders and PMs to identify and address code vulnerabilities before deployment, significantly reducing security risks. For investors, this development signals a growing emphasis on secure coding practices, potentially increasing the attractiveness of software projects that leverage such tools.

GitHub has launched Agentic autofix in public preview for code scanning alerts, automating the remediation process by exploring files, proposing fixes, and validating them with CodeQL. Available to organizations with GitHub Code Security and Copilot licenses, it streamlines fixing alerts into a draft pull request, consuming AI Credits and GitHub Actions minutes during use.

GitHub's launch of Agentic autofix for code scanning alerts automates the remediation process, allowing developers to quickly address vulnerabilities by generating draft pull requests. This reduces the time and effort needed for code maintenance, making it a significant tool for builders and PMs focused on improving software security and efficiency.

OpenAI's GPT-5.6 family, featuring Sol, Terra, and Luna models, is now available in GitHub Copilot, catering to various coding needs from complex reasoning to lightweight tasks. Sol targets high-complexity projects, Terra serves as a balanced option, and Luna focuses on cost-efficient, smaller tasks. Access varies by subscription tier, with a gradual rollout across multiple platforms.

The introduction of OpenAI's GPT-5.6 models in GitHub Copilot allows developers to choose the right AI tool for their specific coding tasks, enhancing productivity and reducing development time. This tiered access can influence project planning and budgeting, making it essential for PMs and investors to consider the cost-benefit of integrating these capabilities into their workflows.

GitHub Copilot now provides high-level overviews of repositories for first-time visitors, summarizing their purpose, technologies, and contribution guidelines. Users can request an overview directly from the repository page or through Copilot Chat, which can also generate a README if one is missing. This feature is available to all GitHub Copilot plans.

GitHub Copilot's new feature that provides high-level overviews of repositories simplifies onboarding for new contributors, enhancing collaboration and reducing the learning curve. For builders and PMs, this means faster project ramp-up times and potentially increased contributions, while investors can see improved project viability through better engagement metrics.

The June 2026 updates for Visual Studio Code (v1.123 to v1.127) enhance GitHub Copilot with features like an integrated browser, parallel sessions, clearer cost visibility, and improved Autopilot functionality, enabling developers to manage tasks more efficiently. These enhancements streamline workflows and provide better tools for agentic development.

The June 2026 updates to GitHub Copilot in Visual Studio Code introduce features like an integrated browser and parallel sessions, which enhance developer productivity and task management. For builders and PMs, these improvements signal a shift towards more efficient development environments, while investors should note the growing integration of AI tools in software development workflows, indicating a potential increase in demand for such solutions.

GitHub's Copilot usage metrics API now includes two new metrics: average pull request review time and review cycles per adoption phase. These metrics help quantify the impact of Copilot on engineering throughput, allowing teams to assess the efficiency of their review processes based on AI adoption levels.

GitHub's Copilot usage metrics API now includes average pull request review time and review cycles per adoption phase, providing teams with quantifiable data on how AI impacts engineering efficiency. This allows builders and PMs to make informed decisions on AI integration, while investors can assess the potential return on investment in AI-enhanced development processes.

Kimi K2.7 is now available for Copilot Business and Enterprise plans, expanding its accessibility beyond Pro tiers. This open-weight model, hosted on Microsoft Azure, offers a cost-effective coding solution but requires administrators to enable it in settings for organizational use.

The release of Kimi K2.7 for Copilot Business and Enterprise plans signifies that advanced AI coding tools are now accessible to a wider range of organizations, potentially reducing development costs. Builders and PMs should consider integrating this model to enhance productivity, while investors might see increased adoption rates in enterprise environments as a signal of growing demand for AI-driven solutions.

The GitHub Copilot app is now accessible to all users across every Copilot plan, including Free and Education. Users can sign in with their GitHub accounts to initiate agent-driven development on macOS, Windows, and Linux, with the option to run sessions using their own model provider without a subscription.

The availability of the GitHub Copilot app to all users, including those on Free and Education plans, lowers the barrier for developers to access advanced AI coding assistance. This democratization of AI tools can accelerate development cycles and enhance productivity, making it essential for builders and PMs to consider integrating such tools into their workflows.

GitHub will retire the Copilot Billing Preview app on August 3, 2026, transitioning users to enhanced billing settings that provide more comprehensive insights into AI usage and spending. Users can now manage budgets and access detailed usage data directly from their GitHub billing settings.

The retirement of the Copilot Billing Preview app and the transition to enhanced billing settings on GitHub will allow builders and PMs to better manage their AI usage and budget, providing detailed insights into spending. This development signals a shift towards more transparent and user-friendly financial management tools for AI services, which is critical for investors assessing cost efficiency.