Artificial Intelligence

GitHub Copilot Adds Natural Language Image Generation and Editing

GitHub has announced that developers can now generate and edit images using natural language directly within GitHub Copilot, eliminating the need to switch between separate tools during a workflow.

Radwan AbdulhadiEditor and technology writer

Published: October 29, 2026 · Updated: October 29, 2026 · 3 min read

GitHub Copilot Adds Natural Language Image Generation and Editing — Image: GitHub Changelog

New Image Generation and Editing Capabilities in GitHub Copilot

GitHub has introduced a new capability allowing users to generate and edit images using natural language directly inside GitHub Copilot. According to the official announcement detailed in the [GitHub Changelog](https://github.blog/changelog/2026-10-28-generate-and-edit-images-directly-in-copilot/), when a user requests an image, their selected reasoning model calls a dedicated image-generation tool powered by GPT Image 2.5 Flare.

Previously, obtaining a visual asset while working in Copilot required developers to step away to a separate tool, describe the requirement again, generate the asset externally, and manually bring it back into their projects. This update removes that context switch, allowing creators to move seamlessly from an idea to a usable image without breaking their coding flow.

Workflow Integration and Workspace Saving

The feature is designed to fit naturally into existing routines. Developers can describe an image conversationally in any Copilot chat session, whether they are starting a brand new request or are right in the middle of a complex coding task.

Once a prompt is submitted, Copilot previews the generated image inline and saves it directly to the user's workspace. This integration streamlines asset creation across supported environments.

Enterprise Controls and Watermarking Options

For enterprise environments and organizational compliance, administrative controls have been included from the start. Organization and enterprise owners can turn off image generation entirely by disabling the image generation model using [Copilot model policies](https://docs.github.com/copilot/concepts/enterprise/policies?utm_source=image-generation-changelog-policy-docs&utm_medium=changelog&utm_campaign=universe-2026).

Additionally, users have the option to opt in to a visible "Made with AI" watermark on images that Copilot generates and edits. This watermarking setting is turned off by default.

Availability and Access Requirements

To test out the new visual features, users can open a Copilot chat in Visual Studio Code, the GitHub Copilot app, or the Copilot CLI, and describe the desired image.

Image generation is currently available in public preview for Copilot Pro, Copilot Pro+, Copilot Business, and Copilot Enterprise subscribers. GitHub notes that image generation is billed separately from regular chat models based on image tokens that count toward a plan's specific AI credits.

Community Feedback and Further Discussion

Developers looking to share feedback, ask questions, or learn more about how other teams are utilizing the new update can join the conversation by visiting the [GitHub Community](https://github.com/orgs/community/discussions/211015?utm_source=image-generation-changelog-community-discussion&utm_medium=changelog&utm_campaign=universe-2026) discussion thread dedicated to this feature.