Figma

Recently shipped parallel AI image edits that run multiple generations at once.

AI image edits no longer block the canvas. Trigger multiple edits from the toolbar and keep designing while they load. No more waiting for one to finish.

GPT-5.6 is now available in Figma Make. Open the model picker in the Make chatbox to select it. Available on all plans.

Design agent gained web search. Fetches live content, stats, and images directly onto the canvas without leaving Figma.

Grid went GA with smarter auto-layout behavior. Reorder columns and rows, auto-position items into empty cells, and auto-add or remove rows to fit.

Tab Groups, larger video uploads, and more. Tab Groups organize desktop tabs. Video uploads grew to 300 MB. The Chrome extension captures webpages as editable layers. Figma Slides added sections. Buzz gained bulk editing. Admins can buy AI credits pay-as-you-go; Enterprise customers have a programmatic API.

Workspace-level web publishing for Make and Sites. Org admins manage web publishing permissions at the workspace level.

The Figma MCP server added Slides, uploaded fonts, and more. Four new capabilities—Slides support, access to uploaded fonts, and two others.

Code Connect 1.4.9 fixed parser paths and template-only builds. Paths with spaces no longer break Swift, Compose, or custom parser invocations. Template-only projects skip the native parser, speeding up CI. Swift syntax support raised to 603.x.

Bug fixes shipped in versions 26.21.0, 26.22.0, 26.22.1, 26.24.0, and 26.25.0.