Google is inviting Gemini Advanced users to try out its new experimental AI model, Gemini-Exp-1206, which it says offers better performance when it comes to coding, math, reasoning, and following instructions.

“Whether you’re tackling complex coding challenges, solving mathematical problems for school or personal projects, or providing detailed, multi-step instructions to craft a tailored business plan, Gemini-Exp-1206 will help you navigate complex tasks with greater ease,” said Google’s Patrick Kane in a blog post.

Eligible users can access the new model through the Gemini model drop-down on desktop and mobile web.

However, the company noted that due to the experimental nature of the new model, there might be some limitations like unexpected responses at times, restricted access to real time information, and possible compatibility issues with other Gemini features.

Last week, Google introduced the Gemini 2.0 Flash model.

“With this new model, Gemini 2.0 will unlock an even more helpful Gemini assistant. Gemini 2.0 Flash Experimental is accessible via the Gemini model drop-down on desktop and mobile web and will be coming to the Gemini mobile app soon,” said Dave Citron, Senior Director, Product Management, Gemini app, in a blog post.

Tech outlet 9to5Google reported on Monday (December 16, 2024) that a model switching option was being rolled out to Android app users, allowing them to try out the Gemini 2.0 Flash experimental model.

People using generative AI tools should be aware that new and experimental models may not always provide answers that are helpful or even accurate.