After announcing the Gemini family of models last week and bringing it to the Bard chatbot experience, Google is now bringing Gemini to developers by launching a series of new and updated services today. One of those services is AI Studio — which was formerly known as MakerSuite.
AI Studio is an online tool for developers that acts a bit like a gateway to the wider Gemini ecosystem, starting with Gemini Pro and then, sometime next year, Gemini Ultra. Using the service, developers can quickly develop Gemini-based prompts and chatbots — and then receive API keys to use in their apps or access the code to edit it in a more full-featured IDE.
It’s important to note that there’s a relatively generous free limit, with up to 60 requests per second, which should be enough to quickly iterate on ideas without facing burdensome limitations, and maybe even enough to power some less-used applications in production .
However, there’s a price to pay here: For developers using the free tier (and that’s pretty much everyone for now, as Google plans to release a paid version only early next year — likely to coincide with the release of of the Gemini Ultra model in GA), Google reviewers can see API and web app input and output to “improve product quality”. However, Google notes that this data is not recognized by the user’s Google Account and API key.
Compared to the previous version of MakerSuite/AI Studio, this updated version looks quite more substantial. Among other things, it will offer support for both Gemini Pro and the Gemini Pro Vision model, allowing developers to work with both text and images (though not to create images).
“We’ve really designed this to be the fastest way to build with Gemini,” Josh Woodward, Google’s VP of Google Labs, told me. “We really want to invite developers to come play with it. It’s the first version and we have a lot of improvements we’re already making now for future updates, but we’re trying to design it in a way that people can just jump in and really start building with it.”
In the web interface, developers can select their models, adjust the temperature to control the creative range of the output, and provide examples to provide guidance on tone and style. You can also tune the model’s security settings. It’s worth noting that in MakerSuite, you could disable all guardrails by telling the system not to block any responses, while in AI Studio, the lowest setting is now “block few”.
There are also different workflows for creating free-form, structured prompts and chat messages.
Woodward noted that the team tried to design AI Studio so that even the free tier doesn’t feel like a trial or closed product. Indeed, assuming the free tier’s rate limits are sufficient for their use cases, developers can start publishing their apps to AI Studio or using them through Google’s API or SDKs right away.
Jeanine Banks, vice president and general manager of Google’s developer X teams and head of developer relations, also emphasized that AI Studio is a gateway to Google’s broader artificial intelligence ecosystem, specifically Vertex AIGoogle’s enterprise AI developer platform.
“[We have] this idea of ’developing with Google’ where you can come in, build something, run it, develop it, let people use it and have this generous free tier. Then we’re also shipping a whole set of SDKs that allow developers to run and build apps with Gemini Pro that can run almost anywhere, from the back-end with support for Node.js and Python, to mobile, with support for Java, Kotlin and Swift and on the web, of course, with JavaScript,” he explained.
He noted that the team wants this transition between moving from AI Studio to Vertex to be as seamless as possible. Woodward added that the strong SDK support came directly from user feedback. “The first version we showed people, they said, ‘I love how easy it is to prompt. Now I want to go to the code.” And there was such a cliff that we had to fill,” he told me.
Speaking about the overall ecosystem, Banks also explained that Google plans to bring Gemini to Chrome Dev Tools and Google’s Firebase mobile development platform early next year.
Given the speed at which genetic AI is developing, it’s hard to even predict which developers will want to use these tools next, but Banks and Woodward emphasized that Google plans to make AI Studio an easy step for developers of all skill levels.
“I hope that AI Studio, in some ways, will not just be seen as a prompting tool or something that only developers go to, but is actually, in some ways, a developer and creativity tool where people can come up with ideas to work with these models and all the features that will come out in the next year,” Woodward said.
