GitHub Copilot Vision generally available for code context
GitHub Copilot Vision is now generally available, allowing users to attach images and PDFs directly to chat prompts for context alongside code. This feature enhances Copilot's reasoning capabilities by enabling it to analyze visual information. It is available across GitHub Copilot Chat in VS Code, github.com, and GitHub Copilot CLI for all Copilot subscribers, with no policy changes required.
- →Copilot Vision now supports images and PDFs in chat prompts
- →Copilot Vision integration across GitHub surfaces
- →Copilot Vision available to all subscribers
- →Data retention for Copilot Vision
Features (1) ›
- Copilot Vision now supports images and PDFs in chat prompts
GitHub Copilot Vision is now generally available, enabling users to attach images (JPEG, PNG, GIF, WebP) and PDFs directly to chat prompts. This allows Copilot to analyze and reason about visual content in conjunction with code.
Enhancements (2) ›
- Copilot Vision integration across GitHub surfaces
The vision capability is integrated into GitHub Copilot Chat within VS Code (via paste, drag-and-drop, or right-click), on github.com, and within the GitHub Copilot CLI. This provides a consistent experience for analyzing visual context across different development environments.
- Copilot Vision available to all subscribers
Copilot Vision is now accessible to all Copilot plans, including Free, Pro, Pro+, Business, and Enterprise. This eliminates the need for specific policy changes or admin actions, making it broadly available.
Notes (1) ›
- Data retention for Copilot Vision
For users on Copilot Business and Copilot Enterprise plans, GitHub retains image and PDF attachments for approximately 24 hours to ensure service delivery. Previously, this feature required the Editor Preview Features policy to be enabled.
https://github.blog/changelog/2026-07-01-copilot-vision-is-generally-available