Hi ! i asked myself if it’s possible to make google research for the present with google gemini integration, and if it’s possible, what version of gemini I need ?
Yes you can actually call the Gemini API and I believe you can use their API for Gemini 2.0 also https://www.make.com/en/integrations/gemini-ai
Hello @Hugo_Barbe,
As I can understand you need to check some live(instant) research with Google Gemini API.
But as of now, there are no specific AI models is support instant search using API.
When you’re accessing using its chat mode(UI mode) that time it uses its background to search on Google Search to get a response of some live results.
During my research, I’ve found a couple of new things to do live research with Gemini API. Just cross-check if it fits your requirements.
As you can see, the 2.0 model do not give me real response, this is not the good answers, how can I have right and instant/real answers ?
1 Like
I have the same question as you.
If I am not mistaken in the advanced settings somehow, in the tools section you have to enable the grounding tool.
I have not figured it out yet. If you did I would appreciate it if you could post how you enabled the functionality.
I am pretty sure you can enable grounding within the API call. The question is, how to do it?
I tried using the tools input and setting it to google_search, but this does not work. Has anyone succeeded using the grounded google search in an Gemini AI API call?
FYI, I’ve resolved this by leveraging exa.ai 's capabilities for web crawling and LLM-based responses, a feature set that addressed the problem better than Gemini.