Skip to content

Gemini for your website

A Gemini chatbot for your website, answering from your own content

Choose Gemini Flash or Gemini Pro as your agent's model, and Google's Gemini writes the replies your visitors get, grounded in your website, files and Q&A. Everything around it, from actions and forms to the inbox and hand-off, works as on any intoCHAT agent. You don't need a Google Cloud account or an API key.

Example: Gemini Flash answering from a bike shop's service page

Visitor: How much is a standard service, and do I need an appointment?

Agent: A standard service costs 79 euros and includes a safety check, brake and gear adjustment, and chain cleaning. You can drop your bike off without an appointment from Monday to Saturday before 11 am, and it is usually ready the next working day.

Visitor: Does that include new brake pads?

Agent: Brake pads aren't included. If yours are worn, we call you before replacing them, and the parts are charged separately.

A strong model still needs your facts and a place to work

Gemini knows a lot about the world, but nothing about your delivery times, your service prices or the exceptions in your returns policy. To turn the raw API into a support bot, you would still have to build the knowledge search, a chat widget for your site, lead and booking forms, a way to hand a conversation to a person and an inbox for your team. intoCHAT is that layer. It indexes your content, hands the matching passages to the Gemini model you picked, and gives you the widget, the actions and the inbox around it.

What you get with Gemini on intoCHAT

Google's models write the replies; intoCHAT supplies your knowledge, the tools and the place your team works.

  • Two Gemini models to choose from

    Gemini Flash, fast and inexpensive at 1 message per reply, is included from Starter. Gemini Pro, built for stronger reasoning, uses 2 messages per reply and is included from Pro.

  • Answers grounded in your knowledge

    Gemini writes each reply from the passages intoCHAT finds in your website pages, files, text snippets, Q&A pairs, Notion pages or Google Drive files, and follows your instructions and guardrails.

  • Every action still works

    API actions, client-side actions, chat forms, procedures, booking, product search and hand-off to your team work with Gemini exactly as they do with the default model.

  • One model setting, every channel

    The model you pick writes the replies in the chat widget, the inline iframe and chat link, the Help Page, Slack and email. Your team follows every conversation in the same Inbox.

  • A fallback when Google is slow or down

    If Google's API fails, or takes more than 25 seconds before the reply starts, GPT-6 Luna answers instead. Your visitor still gets a reply, and it uses one message.

How to set up a Gemini chatbot

  1. Create and train your agent

    Create an agent, add your website, upload files or write Q&A pairs, and train it. Use the instructions to set the tone, the topics it covers and what it says when the answer isn't in your knowledge.

  2. Choose a Gemini model

    On the Setup tab, open the Model card and pick Gemini Flash or Gemini Pro. Chat with it in the Playground before you save, or put it next to GPT-6 Luna or a Claude model in Compare.

  3. Put it where your customers are

    Add the script tag, the WordPress plugin or the Shopify snippet to your site, or share your Help Page. From Starter you can also connect Slack or give the agent its own email address.

Gemini models on intoCHAT

  • Gemini Flash

    Fast tier, 1 message per reply, on Starter, Pro and Enterprise. Quick, inexpensive replies for everyday questions at volume. It is one of the fast models Auto can pick for simple messages, after Claude Haiku 5.5.

  • Gemini Pro

    Advanced tier, 2 messages per reply, on Pro and Enterprise. Stronger reasoning for questions that combine several facts from your knowledge, or several steps.

  • The same search for every model

    Whatever the model, intoCHAT picks up to six passages, about 12,000 characters, from your knowledge for each reply. A Gemini model reasons over those passages, so your answers depend on good sources as much as on the model.

  • Images yes, PDFs no

    Gemini reads images that visitors attach. PDFs are read by OpenAI's models only: with Gemini, the agent is told a PDF was attached and answers from what the visitor wrote.

  • See which model answered

    Why this answer under each reply names the model that wrote it and says when the default model stepped in. The Models used card on the Dashboard counts replies per model.

  • Different models for different agents

    Each agent has its own model setting: a fixed model or Auto. Run Gemini Pro on an agent that answers detailed product questions and Gemini Flash on a busy FAQ agent in the same account.

Frequently asked questions

Can I build a website chatbot with Google Gemini?

Yes. Create an intoCHAT agent, train it on your content and pick Gemini Flash (from Starter) or Gemini Pro (from Pro) as its model. Gemini then writes the replies visitors get in the chat widget on your site. You don't need a Google Cloud account, an API key or any code beyond the embed snippet.

Is intoCHAT made by Google or partnered with it?

No. intoCHAT is an independent product. It uses Google's Gemini models through Google's API, in the same way it uses OpenAI's and Anthropic's models through theirs. intoCHAT isn't affiliated with or endorsed by Google. Gemini and Google are names belonging to Google.

Gemini Flash or Gemini Pro: which should I use?

Start with Gemini Flash for most support questions: it is fast and uses one message per reply, the same as the default model. Move to Gemini Pro when questions need more reasoning, for example comparing products or applying a policy with several conditions. Each Pro reply uses 2 messages. Try both side by side in Compare before you save.

How does Gemini compare with Claude or GPT-6 Luna on intoCHAT?

All of them answer from the same knowledge, with the same instructions, actions and channels, so you can switch at any time. They differ in style, length and speed, and in how many messages a reply uses: Claude Haiku 5.5 (the default), GPT-6 Luna and Gemini Flash use 1, Claude Sonnet 5.5 and Gemini Pro use 2, and Claude Opus 5.5 uses 5. Your own Playground tests are the best guide.

What happens if Google's API is down?

If Google fails before the reply starts, or doesn't respond within 25 seconds, intoCHAT tries GPT-6 Luna once instead. That reply uses one message: the extra message of Gemini Pro is given back. A reply that has already started streaming can't switch models, so in rare cases it may stop short.

Where does visitor data go when Gemini answers?

The visitor's messages, the conversation so far, your instructions and the knowledge passages used for the reply go to Google. intoCHAT's background work, such as searching your knowledge and spam checks, always uses OpenAI, so messages also reach OpenAI. If your privacy policy names the companies that process chat messages, list both.

Can I use Gemini on the Free plan?

No. Agents on Free answer with Claude Haiku 5.5, the default model, or GPT-6 Luna. Gemini Flash is included from Starter and Gemini Pro from Pro, and Starter starts with a 7-day free trial. After a downgrade, replies switch to Claude Haiku 5.5 until you upgrade again or pick another model; your setting is kept.

Put Gemini to work on your own content

Create your agent free, train it on your website, then switch it to Gemini Flash or Gemini Pro on Starter or Pro and try it in the Playground.

Create your agent free

Free plan, no credit card needed.