> ## Documentation Index
> Fetch the complete documentation index at: /docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Google Gemini

> Text, images, speech and video from Gemini. A second opinion on any of them, and the cheapest route to long video.

Text, images, speech and video from Gemini. A second opinion on any of them, and the cheapest route to long video.

## Connecting

In the dashboard, open **Connectors**, find Google Gemini and press **Connect**. The key you enter is checked against Google Gemini before anything is saved, so a key that is refused never becomes a connection that fails later.

### What you will need

| Field | What it is |
| - | - |
| API Key | Your Google Gemini API key from Google AI Studio |

### Getting the credential

1. Go to [Google AI Studio](https://aistudio.google.com/apikey) and create an API key
2. Paste it into the connector
3. Some models, and Veo video, require billing enabled on the Google Cloud project

## Notes

* Images and audio come back as hosted URLs, not base64 — your AI teammate gets a link they can pass on.
* `gemini-3.1-flash-lite-image` supports `image/jpeg` only; ask for PNG on a larger model.
* Image sizes use an uppercase K: `512px`, `1K`, `2K`, `4K`.
* Video is asynchronous. `generate_video` returns an operation; poll `get_video_operation` until `done` is true.

## What your AI teammate can do with it

| Tool | What it does |
| - | - |
| `generate_text` | Generate text with a Gemini model. Use for writing, reasoning, summarising and Q\&A. |
| `generate_image` | Generate an image from a text prompt with a Nano Banana model. Returns a hosted URL. gemini-3.1-flash-lite-image supports image/jpeg only; 4K needs gemini-3.1-flash-image or gemini-3-pro-image. |
| `generate_speech` | Synthesize speech from text. Returns a hosted audio URL. Include delivery notes in the text itself, e.g. 'Say cheerfully: ...'. |
| `generate_video` | Start a Veo video generation. Returns a long-running operation name — poll it with get\_video\_operation until done, then read the video URI from the result. |
| `get_video_operation` | Poll a Veo generation started by generate\_video. The operation name looks like 'models/\<model>/operations/\<id>' — pass the two halves separately. Returns done=false while it runs; when done=true the URI is at response.generateVideoResponse.generatedSamples\[0].video.uri. |

## Useful resources

* [Gemini API docs](https://ai.google.dev/gemini-api/docs)
* [Image generation](https://ai.google.dev/gemini-api/docs/image-generation)
* [Veo video](https://ai.google.dev/gemini-api/docs/veo)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.