Hi! I was wondering if it's possible to perform Google searches in real-time using a Google Gemini integration, and if so, which version of Gemini would be required?
Yes, you can indeed call the Gemini API, and I believe you can use their API for Gemini 2.0 as well: https://www.callin.io/en/integrations/gemini-ai
Hello,
As I understand it, you need to perform live (instant) research using the Google Gemini API.
However, currently, there are no specific AI models that support instant search via API.
When you access it in chat mode (UI mode), it uses its background processes to search Google Search for responses based on live data.
During my investigation, I discovered a couple of new approaches for conducting live research with the Gemini API. Please verify if these align with your requirements.
Please provide the rewritten markdown content; it should be in markdown format.
I'm facing the same issue.
As far as I recall, within the advanced settings, under the tools section, you need to enable the grounding tool.
I haven't managed to get it working yet. If you happen to figure it out, I'd be grateful if you could share how you enabled this functionality.
I'm quite confident you can enable grounding through the API call. The main question is how to implement it.
Just wanted to share that I've found a solution by utilizing exa.ai's web crawling and LLM response features. This approach proved more effective for the issue than Gemini.
Could you create a brief guide for exa and LLMs?
I utilized [{"googleSearchRetrieval": {}}]
within the tools section, and it functions correctly for me.
It appears to be
[
{"googleSearch": {}}
]
currently.