Published Sep 3, 2026
Framer’s Agent now supports 3D transform properties including Rotate X/Y/Z, Perspective, Origin, Depth and Backface while keeping the resulting layers editable on the canvas.
What's New
Framer’s Agent now supports a broad set of 3D Transform properties, including Rotate X, Y and Z, Perspective, Origin, Depth and Backface. The Agent can use those controls while building visual effects directly on the canvas.
Why It Matters
3D web effects often require several connected transform settings that are tedious to tune manually. Giving the Agent access to the same properties can make complex spatial compositions faster to create while still leaving designers with editable layers rather than a flattened result.
What Happened
Framer announced the update on September 1, 2026. The company says 3D work created by the Agent remains built from normal Frames and Effects that can still be selected and customized on the canvas. Framer demonstrated the feature with a logo animation built using Opus 5 Fast with Light reasoning.
What's Next
The important question is how far Framer extends Agent control across motion and spatial design. As more visual properties become agent-accessible, prompts can handle increasingly complex first passes while manual canvas editing remains available for refinement.
TUF Take
The useful part is editability. Generating a 3D effect is less interesting if the result becomes a black box. Framer is keeping the generated structure inside the same canvas system, which makes the Agent feel more like another way to manipulate the design than a separate generator.