Opera Integrates Google Gemini with AI Aria

Last Tuesday, the 28th, the Opera made an announcement about the incorporation of a new LLM into its artificial intelligence integrated into the browser, the Aria. According to the information released, the company chose to adopt the Gemini model to enhance AI capabilities.

The use of Google Cloud it was also announced as essential to enable the efficient processing and functioning of this AI. Currently, Aria with Gemini can only be accessed through the beta version of the browser, but it is expected that it will soon become accessible in other versions.

Read too
Google announces news for its Gemini artificial intelligence
Gemini: Google’s artificial intelligence gains integration with YouTube and Gmail in Brazil

Aria usa ChatGPT e Google Gemini

In the past, when Aria was first introduced to the publicit was based on a version of LLM GPT, developed by OpenAI, the same organization behind well-known systems like ChatGPT and the Copilot.

Opera reported that it has plans to keep Aria versatile, allowing it to use different artificial intelligence models according to the needs of each specific task. It is interesting to note that it already operates with more than one hundred and fifty different LLMs. GPT and now Gemini are just two examples among the best known.

With regard to image creation, Opera indicated that it will adopt Imagen 2, created by DeepMind — which is part of the conglomerate Google. I was willing to try this image creation tool myself, however my browser has not yet received this specific update.

It is worth mentioning that the relationship between Opera and Google is not something recent. The Google search engine has been the default engine in the Opera browser for a long time. Now, with the adoption of the Gemini model, Google joins OpenAI in the foray into the territory of advanced AI.

Aria is still a very inefficient AI

Opera has been my default browser for many years, perhaps more than a decade. And right when Aria was launched, I started testing the tool with some enthusiasm. But, I must say that I ended up disabling the feature because Aria was still very “raw”.

In addition to the delay in generating responses and, occasionally, some mistakes and hallucinations in the responses, Aria still lacks extra features to improve its usability. However, there is hope that implementing the Gemini model will bring substantial improvements.

Despite the initial challenges, it is fair to recognize that Aria offers very useful functionality for understanding the meaning of words in Portuguese. The user simply selects the desired term, requests the definition and Aria will quickly display the meaning, eliminating the need to open a new browser tab.

It is also important to highlight the Opera One and Opera GX browsers, aimed especially at gamers. Both are compatible with Aria. Additionally, Opera has taken strategic steps by establishing a dedicated AI center in Iceland, equipped with sophisticated devices Nvidia DGX.

In relation to Google, after a more discreet start in the field of artificial intelligence, we are now seeing accelerated progress. The Gemini model has already found its way into Galaxy AI and there are predictions that it will soon be introduced in cell phones from the Oppo and OnePlus brands. Honor, another smartphone manufacturer, also announced intentions to integrate this Google AI into its devices.