GitHub Copilot now generates and edits images from natural language prompts

GitHub Copilot now supports generating and editing images using natural language, GitHub announced in a changelog post on October 28, 2026.

According to GitHub, when a user asks for an image, the reasoning model they have selected calls a dedicated image-generation tool. The feed summary of the announcement is truncated at that point, so the name of the model powering the tool is not given here.

What GitHub announced

• Copilot users can create images through natural language requests.
• Existing images can be edited the same way.
• The selected reasoning model decides when to invoke the image-generation tool, which is separate from the reasoning model itself.

This is a product feature for Copilot, not the release of a new model.

What this means

The design routes image requests through the reasoning model the user has already chosen, which then hands the task to a dedicated image tool. Image generation and editing are therefore part of the same natural-language workflow as other Copilot requests, and users do not need a separate interface for them.

Pricing, availability by plan, supported image models and usage limits are not covered in the summary. Check the full GitHub changelog entry at github.blog for those details.