# Update Google Gemini Chat Model node to support new grounding feature

**URL:** <https://community.n8n.io/t/update-google-gemini-chat-model-node-to-support-new-grounding-feature/60864>\
**Category:** Feature Requests\
**Tags:** node, gemini\
**Created:** [November 6, 2024, 5:30am UTC](https://community.n8n.io/t/update-google-gemini-chat-model-node-to-support-new-grounding-feature/60864 "2024-11-06T05:30:12Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![wowsers007](https://dub1.discourse-cdn.com/n8n/user_avatar/community.n8n.io/wowsers007/32/47309_2.png) [@wowsers007](https://community.n8n.io/u/wowsers007)\
**Post date:** [November 6, 2024, 5:30am UTC](https://community.n8n.io/t/update-google-gemini-chat-model-node-to-support-new-grounding-feature/60864/1 "2024-11-06T05:30:12Z")

</div>

## The idea is:

Google just released a cool new feature called grounding that gives Gemini access to Google search, which is awesome. Please update the Google Gemini Chat Model node to support this new feature.

## My use case:

Academic assistant.

## I think it would be beneficial to add this because:

Grounding solves several problems:

1. Access to recent data (e.g. who won the 2024 MLB world series. Gemini doesn’t know the answer, but Google search does.
2. Reduces hallucinations by citing specific sources.

## Any resources to support this?

> **[Gemini generateContent API  |  Google AI for Developers](https://ai.google.dev/gemini-api/docs/google-search)**
>
> Ground your model's responses in real-time information from Google Search to improve factual accuracy and provide citations.

## Are you willing to work on this?

I’m not a developer, sorry.

---

<div class="post-metadata">

**Author:** ![hink](https://dub1.discourse-cdn.com/n8n/user_avatar/community.n8n.io/hink/32/49259_2.png) [@hink](https://community.n8n.io/u/hink)\
**Post date:** [December 13, 2024, 8:44am UTC](https://community.n8n.io/t/update-google-gemini-chat-model-node-to-support-new-grounding-feature/60864/2 "2024-12-13T08:44:10Z")

</div>

I am not a node developer, and not necessarily wanting to build a custom version, but I believe I have either completed or given a good start to someone that wants to test adding grounding in a custom build and submit a pull request.

Google Documentation: [Grounding with Google Search &nbsp;|&nbsp; Gemini API &nbsp;|&nbsp; Google AI for Developers](https://ai.google.dev/gemini-api/docs/grounding?lang=python)

> <https://gist.github.com/hink/c4c1ecb91f938272208adf5c0f1d298f>

---

<div class="post-metadata">

**Author:** ![Devilish](https://avatars.discourse-cdn.com/v4/letter/d/bc8723/32.png) [@Devilish](https://community.n8n.io/u/Devilish)\
**Post date:** [February 28, 2025, 3:50am UTC](https://community.n8n.io/t/update-google-gemini-chat-model-node-to-support-new-grounding-feature/60864/3 "2025-02-28T03:50:40Z")

</div>

This would be awesome! Could easily cut the API usage for google search with this.

---

<div class="post-metadata">

**Author:** ![yos233](https://dub1.discourse-cdn.com/n8n/user_avatar/community.n8n.io/yos233/32/109790_2.png) [@yos233](https://community.n8n.io/u/yos233)\
**Post date:** [August 14, 2025, 8:39pm UTC](https://community.n8n.io/t/update-google-gemini-chat-model-node-to-support-new-grounding-feature/60864/4 "2025-08-14T20:39:05Z")

</div>

Seems like there’s been no progress for the last few months. Is there a timeline for the SaaS version to upgrade and support this natively?

I haven’t tried the http workaround yet, is it good enough to substitute basic queries? I don’t even need history retention, a “fresh” chat each time is fine.

---

<div class="post-metadata">

**Author:** ![Tim\_Yeung](https://dub1.discourse-cdn.com/n8n/user_avatar/community.n8n.io/tim_yeung/32/44775_2.png) [@Tim\_Yeung](https://community.n8n.io/u/Tim_Yeung)\
**Post date:** [December 10, 2025, 1:41am UTC](https://community.n8n.io/t/update-google-gemini-chat-model-node-to-support-new-grounding-feature/60864/5 "2025-12-10T01:41:09Z")

</div>

I have made one that support google search and cachedContent

[https://www.npmjs.com/package/n8n-nodes-google-vertex-cached](https://www.npmjs.com/package/n8n-nodes-google-vertex-cached "https://www.npmjs.com/package/n8n-nodes-google-vertex-cached")
