Figma Make Now Supports Claude Sonnet 4.6 and Multi-Model Switching.

Figma has added Claude Sonnet 4.6 to its Make tool and introduced the ability to switch between different AI models directly within the interface. The update, announced on LinkedIn, gives designers flexibility to choose the most appropriate model for their specific tasks without leaving the Figma environment.

Figma Make, the company's AI-powered design generation tool, previously locked users into a single model per session. The new multi-model switching capability addresses a common workflow friction point, particularly for teams experimenting with different AI approaches or optimizing for specific use cases like UI generation, copywriting, or prototyping.

The addition of Claude Sonnet 4.6 represents Figma's continued investment in offering multiple AI provider options within Make. Anthropic's latest Sonnet variant has gained traction among developers and designers for its improved reasoning capabilities and longer context windows compared to earlier versions.

Figma has steadily expanded Make's AI capabilities since its launch. The tool recently gained voice-to-text input and version history, making it easier to iterate on AI-generated designs. The company has also added drag-and-drop support for reference images, allowing designers to guide AI output with visual context.

Users can now toggle between available models mid-workflow, letting them compare outputs or leverage specific model strengths. A designer might start with one model for initial concept generation, then switch to Claude Sonnet 4.6 for refining detailed component logic or accessibility considerations.

The update arrives as AI-powered design and prototyping tools proliferate. Lovable, v0, and other platforms have incorporated multi-model support to give users more control over AI-generated code and interfaces. Figma's approach keeps model selection within the same design file, reducing context switching.

Figma Make competes in a rapidly evolving category that includes Google's experimental Stitch tool and emerging platforms focused on vibe coding workflows. As we previously tested, these tools differ significantly in their approach to translating natural language into functional designs, with model choice often determining output quality and style.

The ability to switch models addresses a practical constraint: no single AI model excels at every design task. Some generate cleaner component hierarchies, others handle complex layouts better, and some produce more accessible markup. Giving designers model-level control lets them optimize for the specific problem they're solving rather than accepting one-size-fits-all output.

Figma has not disclosed the full list of available models in Make or whether additional providers will be added. The company continues to release incremental updates to Make through its regular release notes sessions, including recent teasers about Vibe Coding enhancements.

For teams already using Figma Make, the update requires no migration or configuration changes. Model switching is available immediately within existing Make projects.