Google Gemini

Connect a Google account to use Gemini models in your processes. metamorphOS calls Google’s Gemini models to generate content from a prompt, create text embeddings, count the tokens a prompt would use, and list the models your key can access.

What you can do

Generate content: send a text prompt (optionally with extra text and file or image attachments) and get back the model’s response.
Embeddings: turn a piece of text into an embedding vector for search or similarity work.
Count tokens: check how many tokens a prompt would use for a given model before you send it.
List models: see which Gemini models your key can use and what each one supports.

Before you start

You need a Google account and access to Google AI Studio, where you create the API key. Creating a key attaches it to a Google Cloud project; you can use an existing project or let AI Studio create one for you.

Set up Google Gemini

Open Google AI Studio. Go to Google AI Studio at https://aistudio.google.com and sign in with your Google account.
Create an API key. Open Get API key, then Create API key, and select or create a Google Cloud project to attach it to.
Copy the key. Copy the key right away and store it somewhere safe.

Connect in metamorphOS

Activate the Google Gemini connector from the marketplace, add an account, and provide:

API key: the key you created in Google AI Studio. Each account can use a different key, so you can point separate accounts at different Google Cloud projects.

Good to know

The model you name in a process (for example, gemini-1.5-flash, gemini-1.5-pro, or text-embedding-004) must be one your key’s project can access. Use List models to confirm what is available.
Rate limits and free-tier quotas are set by Google for each project, not by the connector.