Gemini integration
Use this guide to connect Gemini models to ReplyRabbit for a FreeScout mailbox. In ReplyRabbit settings, this is configured through the Google AI Studio provider.
Looking for the overview?
Overview
ReplyRabbit supports Google AI Studio as a cloud AI provider. After you select Google AI Studio, paste your API key, and choose a model, ReplyRabbit can use Gemini-backed generation for FreeScout drafts and suggestions.
ReplyRabbit remains the FreeScout workflow layer. It provides company context, response style, examples, and reviewable output for agents.
When Google returns cached-token metadata for a Gemini request, ReplyRabbit records it with the usage data. Mailbox token budgets still count total input and output tokens.
Prerequisites
Before you start, make sure you have:
- A working FreeScout installation.
- ReplyRabbit installed and activated.
- ReplyRabbit Pro or Team, which unlocks all AI providers.
- A Google AI Studio account.
- A Google AI Studio API key.
If you are still choosing a provider, read Choosing an AI provider.
Permissions and API keys
ReplyRabbit uses your Google AI Studio API key only when the mailbox provider is set to Google AI Studio. Saved keys are stored as secrets and shown masked after saving.
If the key is rotated or removed, update the mailbox settings before generating more drafts.
Connect Gemini through Google AI Studio
Open ReplyRabbit settings
In FreeScout, open the mailbox, go to Settings, and choose ReplyRabbit.Choose Google AI Studio
In the AI Provider card, select Google AI Studio.Paste the API key
Paste your Google AI Studio API key into API Key.Choose a model
Choose the latest option, a configured Gemini preset, or Custom if your team has a specific provider-supported model ID.Save and test
Save settings, open a FreeScout conversation, and generate a test draft. Review quality before using it on a busy mailbox.
Configuration notes
- Provider label: use Google AI Studio in ReplyRabbit settings.
- Model label: choose a Gemini model option or enter a custom supported model ID.
- Prompt caching: no manual setup is required. ReplyRabbit records provider cache metadata when it is present.
- Budget: set a monthly token budget if you need a hard stop.
- Context: keep company context and response style accurate, since those fields shape the generated draft.
Use cases
- Use Gemini models for FreeScout draft replies.
- Test Google AI Studio in one mailbox before changing another provider.
- Compare Gemini quality and latency against other ReplyRabbit provider options.
Troubleshooting
| Symptom | What to check |
|---|---|
| Google AI Studio is disabled | Confirm the mailbox is on ReplyRabbit Pro or Team. |
| API key fails | Re-copy the key from Google AI Studio and check provider account access. |
| Model error | Try the latest option, or confirm the custom model ID is supported by Google AI Studio. |
| Drafts are too short or off tone | Review company context, response style, examples, and the selected model. |
| Provider cannot be reached | Confirm your server can reach the Google AI Studio API endpoint. |
For broader fixes, see Troubleshooting ReplyRabbit.
Privacy and security notes
When Google AI Studio is selected, ReplyRabbit sends Google the prompt context needed for the AI feature. That can include customer conversation text and mailbox context.
Use a local provider such as Ollama if your policy requires customer content to stay on your own hardware.
Next steps
- Review the Gemini integration for FreeScout page when you need the workflow and buying overview.
- Compare provider tradeoffs in Choosing an AI provider before changing a production mailbox.
- Continue with How to set up ReplyRabbit if company context, examples, or budgets still need attention.
FAQ
Is Gemini listed as Gemini or Google AI Studio?
In ReplyRabbit settings, choose Google AI Studio. That provider includes Gemini model options.
Do I need a Google AI Studio API key?
Yes. ReplyRabbit uses the API key you provide in the mailbox settings.
Can I choose a specific Gemini model?
Yes. Use a configured model option or the Custom field for a provider-supported model ID.
Does ReplyRabbit configure Gemini prompt caching?
There is no manual Gemini cache setup in ReplyRabbit. ReplyRabbit records cached-token metadata when Google returns it.
Does customer data go to Google?
If Google AI Studio is selected, ReplyRabbit sends Google the prompt context needed for the AI feature.