Google Gemini

The Google Gemini provider sends generation requests to Google’s Gemini models through Google’s OpenAI-compatible endpoint. It leaves your computer: prompts, retrieved excerpts of your documents and the scenario so far are sent to Google for every stage routed to it. Use project routing to keep sensitive projects on local models. Before anything is sent, the app shows the request and asks for your approval, and it replaces names, phone numbers, emails and addresses with placeholders by default: see What goes to cloud models.

1. Get an API key

Create a key in Google AI Studio under Get API key. Some models have a free tier with low limits; check the current terms on Google’s pricing page before you rely on it, because Google may use content sent on a free tier to improve its products. A paid account is the safer choice for exercise material.

2. Add the provider

1. Providers → Add provider → Type: Google Gemini. The address https://generativelanguage.googleapis.com/v1beta/openai/ and a suggested model are filled in.
2. Paste the key into API key. It is required.
3. Keep the suggested Default model or pick another. After the provider is saved, the model field offers the models your key can use.
4. Add provider, then Test on the card. The test asks for your list of models, not for a generated answer.
5. In Routing, assign the tasks you want Gemini to run.

Embeddings

Gemini can also index your documents for search. In Routing, give the Embeddings row the Gemini provider and the model gemini-embedding-001, then choose Configured provider under Settings → Embeddings. Your document text is sent to Google while it is indexed. Changing the embedding provider means the index must be rebuilt: use Rebuild index on the project’s Documents page.

Images

Gemini can draw the photographic scene images for injects. Text-bearing media such as posts, tickets, emails and boards are always drawn from templates, so their text stays exact.

1. Edit the Gemini provider, open Advanced and enter an Image model, for example gemini-2.5-flash-image. Check Google’s model list for the current names.
2. In Routing, give the Image row the Gemini provider. The model you choose on that row is used instead of the one in Advanced.
3. Optionally set Price per image, voice or video call in Advanced, so the app can estimate the cost of a batch.

Each image prompt is shown for approval and redacted before it is sent, and the app asks you to confirm a batch of paid images before it starts. Google’s endpoint ignores some settings, such as the exact picture size, so images may come back in a different shape than the scene asked for. If Google’s safety filters block a prompt, the app says so; reword the image prompt and try again.

How the key is stored

The key is encrypted with the operating system’s keychain and kept only on this computer. It is never written to the settings file, a project folder or a log, and it is sent only to Google’s Gemini host. After saving, the form shows only that a key is set, with Replace and Remove.

What it costs

Google charges per token on a paid account. The app adds no cost of its own and does not know current prices. To see an estimated cost on each stage, open Advanced in the provider form and enter the Input price and Output price in US dollars per million tokens from Google’s price list. Without them, stages show their tokens and say the price is not set.

Troubleshooting

Symptom	Fix
Test says the key is not valid	The key is wrong or was deleted in AI Studio. Edit → Replace the key.
“Model not found”	Google retires older models. Open the model field and choose one from the list.
Rate limit or quota errors	Free tiers allow few requests per minute. Wait, route fewer tasks to Gemini, or enable billing.
An answer is blocked or empty	Google’s safety filters refused the content. Rephrase the prompt or use another model for that stage.