Google has released a beta version of Opal, a tool that can  describe, create, and share AI mini-apps. Google describes Opal as a tool to accelerate prototyping AI ideas and workflows, demonstrate a proof of concept with a functional app, and build custom AI apps to boost your productivity at work.

The name Opal is already in use at Google for Opal Security, which is a centralized authorization platform that supports integrations with Google Groups and Google Workspace, but the new beta tool is completely separate.  

The idea is that anybody can use Opal to create AI mini apps that chain together prompts, models, and tools using natural language and visual editing. 

The process is based around the creation of workflows, the steps required to accomplish a specific goal.  

Opal uses a number Google's AI models depending on what the task is, including Gemini 2.5 for creating written output such as blog posts, Imagen 4 for generating images, Veo 3 for creating videos, and AudioLM for audio generation. 

The way Opal works is that you're first asked to describe what you want your app to do - for example, create a blog post with a video. Opal then asks questions such as 'what is the main topic for your blog', and based on the answers creates a visual workflow - what we used to call a flowchart back in the olden days. You can then edit the prompts, add in new steps, and rearrange the flow. 

Once the flow fits your plan, you can share it as an app for people to use using their own Google account.

Behind the scenes Opal comes with a demo gallery with starter templates. You can use these pre-built AI apps as they are, or remix them to better fit your needs. The templates include the blog post writer, social media post, book recommendations, and generating a playlist. More business oriented templates include a business profiler, product research, and video marketer. The other templates can be used to turn a photo into a claymation video, and creating a simple game. 

Does this do more than old style assistants? It's slicker than Microsoft Office Clippy used to be but in reality this is more a demo than a product.  

Elle Zadina, product manager at Google, said in the introductory video: 

"We want to deliver a product that gives users more control and transparency over combining all the capabilities of Google models, without having to code."  

The release came a day after GitHub, which is owned by Microsoft,  released a public preview of Spark, a similar app-building platform. 

More Information

Opal Website

Related Articles

Google Introduces Gemini CLI Open-Source Agent

Google Introduces Gemini Canvas

Gemini On-Device - Generative AI For Robots

Google Adds Open-Source Development Kit To Vertex AI

To be informed about new articles on I Programmer, sign up for our weekly newsletter, subscribe to the RSS feed and follow us on Twitter, Facebook or Linkedin.

Comments

Make a Comment or View Existing Comments Using Disqus

or email your comment to: comments@i-programmer.info