Grid of glass facade panels reflecting blue sky, clouds and diagonal concrete staircases with metal railings, creating overlapping layers of depth
Grid of glass facade panels reflecting blue sky, clouds and diagonal concrete staircases with metal railings, creating overlapping layers of depth

•

7 mins read time

Figma 3D Transforms for Designers: Depth, Motion and Code

Figma 3D transforms arrived in open beta on 12 November 2026. Rotate any layer in space, set perspective on a frame, let the agent animate the scene and hand it off as CSS. Here is what shipped and how to try it.

Ayoub Kada

•

7 mins read time

Figma 3D Transforms for Designers: Depth, Motion and Code

Figma 3D transforms arrived in open beta on 12 November 2026. Rotate any layer in space, set perspective on a frame, let the agent animate the scene and hand it off as CSS. Here is what shipped and how to try it.

Ayoub Kada

Depth used to be something you faked in a mockup. Now it is an editable property of your layout, and designers who learn to think in camera, pivot and depth will direct how interfaces move.

What Figma 3D transforms change for designers

Figma released Figma 3D transforms on 12 November 2026, in open beta for Full seat users. Figma 3D transforms are a native, non-destructive effect: you rotate any layer on the X, Y and Z axes with on-canvas handles or typed values, push it closer or farther with Depth, and set Perspective on a parent frame so its children share one camera. The Figma agent can build and animate these scenes from a prompt, Figma Motion can keyframe them, and Dev Mode hands the result to engineering as CSS, according to Figma's release notes.

The short answer for designers: this is the first time spatial layout lives inside the same file as your components, tokens and prototypes. You no longer fake depth in a plugin, export a flat mockup, or rebuild a card tilt in a separate 3D tool. Use it for tilted cards, layered hero scenes, product reveals and depth in motion. Do not use it as a replacement for real 3D modelling, because it moves flat planes in space; it does not create meshes.

What Figma shipped

Here is what Figma states in its help article on creating 3D effects in Figma Design and the 12 November launch post:

  • Where it lives. Select a layer, click the plus icon in the Effects section and choose 3D transform. Text, fills (including shader fills) and layers stay editable underneath.

  • Controls. Rotation on X (tilt forward or back), Y (tilt left or right) and Z (spin in the canvas plane), Depth along the Z axis, an origin point for the pivot, and a Backface toggle that shows a mirrored front or hides the back when it turns away.

  • Perspective. Applied to a parent frame in pixels. A smaller value makes the effect stronger, a larger one makes it subtler, and it does nothing unless nested layers carry a 3D transform.

  • Keep nested layers 3D. Puts children in one shared space so they can intersect and sort by position. While it is on, opacity is disabled, blend modes, shaders, other effects and clip content are unavailable, and existing fills and strokes on that frame are hidden. "Turn into flat plane" ends the shared space.

  • Agent and Motion. The agent can apply and edit 3D transforms, add motion, or both from one prompt. In Motion you can keyframe Rotate X, Rotate Y, Depth and perspective.

  • Handoff. Dev Mode maps Depth to translateZ, the nested-layer setting to transform-style and Perspective to perspective. A coding agent connected to the Figma MCP server can implement the effect.

  • Access. Using it requires a Full seat and edit access to the file. Figma's launch post describes it as open beta.

Figma's own motion team explains how they used it in a launch video, in a post titled How 3D transforms and Figma's agent brought a launch video to life. The pattern in that post is worth copying: the agent handles repetitive spacing and geometry, while the designer makes the composition calls.

How to try it this week: the Plane, Camera, Motion, Code workflow

1. Plane: start with one flat layer

Pick a real component, such as a pricing card or an app screen frame. Add the 3D transform effect and rotate it about 15 degrees on Y. Move the origin point and watch how the pivot changes the feel. This is the fastest way to learn what each axis does.

2. Camera: set perspective on the parent

Wrap two or three cards in a frame and set Perspective there. Lower the value until the scene feels dramatic, then back off. Turn on Keep nested layers 3D only if layers need to intersect, and remember the trade-off: no opacity, no shaders and no clipping on that frame while it is on.

3. Motion: let the agent do the math

Ask the agent for something repetitive. Figma's example prompts include "Arrange these layers as the faces of a cube and animate it rotating." Then open the timeline and adjust the keyframes it created. Treat the agent's output as a first pass, not a final animation.

4. Code: check the CSS before you promise it

Open Dev Mode on the 3D layer and copy the CSS. If the effect relies on intersecting layers or heavy depth, talk to engineering before it goes into a sprint. A tilted card is easy to ship. A twenty-layer helix on a mobile landing page is a performance conversation.

3D transforms vs the old ways of faking depth

Approach

Where it lives

Editable later

Animated

Handoff to code

Best for

Static perspective mockup (plugin or skewed image)

Flattened layers in Figma

Only by regenerating

No

Designers describe it in notes

Quick presentation shots

Separate 3D tool (Spline, Blender, Cinema 4D)

Outside the design file

Yes, in that tool

Yes

Export video or a runtime scene

True 3D models and lighting

Hand-coded CSS 3D

In the codebase

By developers only

Yes

Already code

Production polish after design

Figma 3D transforms

Native effect on any layer

Yes, non-destructive

Yes, with Figma Motion and the agent

CSS in Dev Mode, MCP for coding agents

UI depth, tilted cards, layered hero scenes

The verdict: use Figma 3D transforms for interface depth that has to stay in sync with your components and ship as CSS. Keep a dedicated 3D tool for real geometry, materials and lighting.

Hype vs reality

The hype says Figma now does 3D. Part of that is true. Rotating, layering and animating planes in perspective, with an agent that can arrange a cube or a ring for you, used to need a separate tool and a handoff. Now it sits next to your design system, and the CSS mapping is clear.

What does not hold: this is 2.5D. You are moving flat layers through space, not modelling objects, so there are no meshes, materials or real lighting. The shared 3D mode switches off opacity, blend modes, shaders and clipping on that frame. It is an open beta, so expect changes. And a scene that looks smooth on a desktop canvas can still be heavy in a mobile browser. Plan the production version, not only the mockup.

Where this takes design work

Over the next one to three years, depth stops being a decoration added at the end and becomes part of the layout system. Spatial interfaces, layered product storytelling and motion that explains state all need designers who can think in a camera, not only a grid. That links directly to spatial UI design and to motion and 3D web design in Morocco, where depth and movement are becoming part of how products present themselves.

It also continues a clear pattern at Figma. The same agent builds Figma shaders and drives Figma Motion animation styles. Effects, motion and code are converging on one canvas, and the designer's job shifts toward directing behavior and judging the result.

Frequently asked questions

Will Figma 3D transforms replace 3D and motion designers?

No. They replace some routine work: perspective mockups, simple card tilts and repetitive spacing along the Z axis. They do not model, light or texture real objects, and they do not replace the judgment that makes motion feel right. Motion and 3D designers who learn the feature gain a faster way to bring their ideas into product files that engineering can read.

Who can use Figma 3D transforms?

Figma says the feature is available on all plans, but using it requires a Full seat and edit access to the file. The launch post describes it as an open beta for Full seat users.

Can the Figma agent build 3D designs for me?

Yes. According to Figma, the agent can apply and edit 3D transforms, add motion, or do both from one prompt, for example arranging layers as cube faces and animating the rotation. Review the result on the canvas and adjust the keyframes it creates.

How do developers build a Figma 3D design?

In Dev Mode, select the layer to copy its CSS. Figma maps Depth to translateZ, perspective to the CSS perspective property, and the nested-layer setting to transform-style. A coding agent connected to the Figma MCP server can also implement the effect.

Are Figma 3D transforms overhyped?

A little. Calling it 3D invites comparisons with full 3D tools that it does not win. As a native way to add depth to interfaces, keep it editable and hand it off as CSS, it is a real improvement.

Am I falling behind if I have not learned 3D yet?

No. Spend an hour this week tilting one real component, setting perspective on its parent and asking the agent to animate it. Thinking in camera, depth and pivot is the skill that carries forward, and this is the cheapest place to learn it.

Getting help with depth, motion and product visuals

Depth only works when it serves the product story. If you want spatial layouts, launch motion or a visual system that stays consistent from canvas to code, DIGCY's motion design and AI video and UI/UX design services can help. DIGCY is a Dribbble Selected Agency and a product design agency in Casablanca working with clients globally.

Let’s keep in touch.

Discover more about high-performance web design. Follow us on Twitter and Instagram.

Building an AI product?

By submitting, you agree to our Terms and Privacy Policy.

Let's Talk

Tell us about your AI product. After you send this, you can pick a 20-minute slot on our calendar. Prefer email? hello@digcy.com We reply within one business day.

Building an AI product?

By submitting, you agree to our Terms and Privacy Policy.

Let's Talk

Tell us about your AI product. After you send this, you can pick a 20-minute slot on our calendar. Prefer email? hello@digcy.com We reply within one business day.

Building an AI product?

By submitting, you agree to our Terms and Privacy Policy.

Let's Talk

Tell us about your AI product. After you send this, you can pick a 20-minute slot on our calendar. Prefer email? hello@digcy.com We reply within one business day.