Google Introduces Gemini Canvas Tool to Convert Sketches into App Code

What Is New

Google has introduced a tool called Gemini Canvas. According to the Google Gemini Blog, this tool allows users to turn hand-drawn sketches into functional, interactive applications [1]. Users can upload their drawings to the Canvas tool to generate and refine code instantly [1].

Separately, developers have reported changes regarding API access. According to the Google AI for Developers forum, the Gemini API will stop accepting requests from unrestricted API keys starting June 19, 2026, to prevent unauthorized usage and billing risks [2]. Developers are advised to either restrict existing keys to the Gemini API via Google Cloud Credentials or generate new restricted keys in Google AI Studio [2].

How to Use It

To use the new feature for app creation, the following steps are provided by Google:

• Create a sketch or drawing of an app idea on paper [1].
• Upload the drawings to the Gemini Canvas tool [1].
• Use the tool to generate and refine the application code [1].

Limits, Privacy and Academic Integrity

Users may face technical limitations regarding model availability. A developer reported that the gemini-2.5-flash-preview-09-25 model was shut down [2]. While one user noted their app using that specific API was still working despite the scheduled shutdown, others reported difficulties finding a working replacement model within the Canvas environment [2]. Additionally, users with paid monthly Gemini subscriptions have noted issues working within the Canvas environment using certain models [2].