Google models power Figma Make

This partnership turns Figma from an app that adds AI features into an app that can swap in better creative models as fast as Google ships them. In practice, that matters because Figma Make is now doing real production work inside large accounts, and faster image generation keeps designers, marketers, and product teams inside Figma instead of bouncing out to separate AI image tools or prompt apps.1 2 3

• Google is supplying core model infrastructure, not just a logo partnership. Figma integrated Gemini 2.5 Flash, Gemini 2.0, and Imagen 4, and early use cut Make Image latency by 50%, which is the difference between AI feeling like live editing and AI feeling like a background batch job.1 2
• This fits Figma’s bigger expansion play. The company added Make, Sites, Draw, and Buzz at Config 2025, then saw more than half of customers spending over $100K ARR using Make weekly by Q4 2025. The Google tie in gives that broader surface area a shared model layer across prototyping, publishing, and asset creation.3 4
• The closest comparable is Canva, which also moved beyond a single editor by combining owned and partner AI models with app level workflows. The difference is that Canva built deeper vertical AI around marketers and templates, while Figma is wiring frontier models into collaborative product creation where the file is already the team’s working canvas.5 6 7

The next step is straightforward. Figma will keep pulling more creation steps, from mockups to images to motion to publishable sites, into one browser workflow. As model quality keeps rising, the durable advantage will be less about having AI, and more about owning the place where teams already edit together, approve work, and ship it.3 6 7