Google AI Studio Lets Developers Ship Apps Under Custom Domains for Free Google AI Studio now lets developers attach a custom domain to apps built in Build mode, replacing the default ai.studio subdomains and generated Cloud Run URLs, with domain connection itself costing $0. Apps remain hosted on Google Cloud Run with automatic SSL, and the Starter Tier still allows two free full-stack apps without a billing account; developers configure DNS records inside AI Studio's publish panel rather than the Cloud Console, and the site goes live once DNS is verified and a TLS certificate is provisioned. Google AI Studio Lets Developers Ship Apps Under Custom Domains for Free Google AI Studio now lets you attach your own branded domain to apps built in Build mode, moving beyond the default ai.studio subdomains. - Google AI Studio Build now supports attaching your own custom domain to published apps. - Previously limited to generated Cloud Run URLs or shared ai.studio subdomains https://ai.google.dev/gemini-api/docs/aistudio-deploying?utm source=alphasignal . - You can connect the domain before publishing, with the site going live once DNS is verified. - Apps are deployed on Google Cloud Run under the hood, with automatic SSL. - Starter Tier still allows two free full-stack apps without a billing account. - Configured entirely inside the AI Studio publish panel https://aistudio.google.com/apps?utm source=alphasignal , no Cloud Console detour. Google AI Studio now lets developers connect custom domains to apps created in Build mode https://aistudio.google.com/apps?utm source=alphasignal . An app can run at an address such as app.example.com while Google continues to host it on Cloud Run. The integrated setup gives client projects, startup experiments, and public tools a branded URL without requiring a separate deployment platform. Google AI Studio’s earlier Custom URLs feature assigned published apps a free address such as your-app-name.ai.studio , as described in the deployment guide https://ai.google.dev/gemini-api/docs/aistudio-deploying?utm source=alphasignal . These globally unique names are allocated on a first-come, first-served basis and return to the available pool when an app is deleted. Two URL paths | Option | Setup | Best fit | |---|---|---| | ai.studio subdomain | Choose an available name when publishing | Internal tools, demos, and short-lived prototypes | | Custom domain | Add Google’s records through a DNS provider | Client work, public apps, and products using an established brand | DNS without the Cloud Console Build mode deploys applications to Google Cloud Run, which provides the underlying domain-mapping capability. The previous manual process required developers to open Google Cloud Console, locate the correct service, verify domain ownership, create a domain mapping, and add records such as a TXT verification record and a CNAME pointing to ghs.googlehosted.com . Cloud Run domain mapping has also been limited by deployment region. An app running in an unsupported region could require redeployment or a Firebase proxy, as developers described in a forum thread https://discuss.ai.google.dev/t/custom-domains-no-longer-possible/168546?utm source=alphasignal . The updated publishing panel brings domain configuration into AI Studio and supplies the required DNS records. Developers can configure the domain before the first publication, according to the available release notes https://getpinnacle.ai/updates?utm source=alphasignal . Google verifies the records and provisions a TLS certificate before serving the app through the branded address. Connect a domain in four steps 1. Open the app in Build mode and select Publish to view its deployment settings. 2. Enter the intended hostname, such as app.example.com , in the custom-domain field. 3. Copy Google’s DNS records into the domain provider’s control panel without removing unrelated records. 4. Wait for DNS verification and certificate provisioning, then test the HTTPS address. DNS changes may take time to reach resolvers, depending on the provider and each record’s time-to-live setting. The records must remain in place after activation so traffic continues to reach the Cloud Run service. The bill has three parts | Item | Cost | Details | |---|---|---| | Domain connection | $0 | AI Studio does not charge a separate fee for attaching the domain. | | Domain registration | Usually $10 to $15 per year | Prices vary by registrar and top-level domain. Published registration estimates https://kah-explorer.com/blog/articles/article-detail/make-your-blog-professional-bind-a-custom-domain-to-google-ai-studio/en/?utm source=alphasignal place common .com and .dev domains in this range. | | Hosting and model usage | Usage-dependent | Eligible accounts can use the Starter Tier https://ai.google.dev/gemini-api/docs/aistudio-deploying?utm source=alphasignal to publish up to two full-stack apps without adding a billing account. Standard Cloud Run and Gemini API charges can apply beyond the included allowances. | Where a domain pays off The free ai.studio address remains suitable for internal tools, hackathon entries, and prototypes shared with a small group. A custom domain becomes useful when the URL needs to remain stable, match an existing brand, or fit into a broader web property. - A client-facing app needs an address independent of Google’s shared namespace. - A product should run under an existing subdomain such as app.example.com . - A startup experiment needs a consistent address for marketing, analytics, and search indexing. - A team wants to retain its Cloud Run deployment rather than move solely for a branded URL. Production checks still apply Custom-domain support covers routing, ownership verification, and TLS. Public deployments still require authentication, authorization, server-side secret storage, logging, error monitoring, privacy review, quotas, and cost controls. Teams should also test Cloud Run scaling, request timeouts, and Gemini API limits before sending sustained traffic to the app.