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

# Google Gemini Setup

> Connect Google Gemini to MagicAds — the broadest provider, powering the Copy, Image and Video studios plus the Fashion Studio and Product Photoshoot plugins from a single key.

## Introduction

Google Gemini is the **broadest and most important** provider on the platform. A single key drives:

* **Copy Studio** — the Gemini text models,
* **Image Studio** — Google Nano Banana 2 (Gemini 3.1 Flash Image),
* **Video Studio** — Google Veo 3.1 and Veo 3.1 Lite, and
* the **Fashion Studio** and **Product Photoshoot** plugins, which use Gemini as their image/video engine.

Because so much depends on it, the Gemini key is the one most worth configuring first.

## How keys are stored

All AI provider keys live in the **AI Settings** screen and are stored **encrypted** in the database. There are no `.env` edits — keys are injected into the drivers at runtime. Users never supply their own key; every generation runs on the admin's key.

<Note>
  A model only becomes available to users when **both** are true: the Gemini API key is saved **and** the model is **enabled** in the vendor modal. The **same** key powers Gemini text, Gemini image, Veo video, and the two plugins — you only enter it once.
</Note>

## Before you start

* Admin access to your MagicAds platform.
* A Google account with access to [Google AI Studio](https://aistudio.google.com/).
* Billing enabled on the Google project if you'll use the paid models (notably Veo video).

## Part 1 — Get your Gemini API key

<Steps>
  <Step title="Open Google AI Studio">
    Go to [aistudio.google.com/apikey](https://aistudio.google.com/apikey) and sign in with your Google account. Accept the terms of service on your first visit.
  </Step>

  <Step title="Create an API key">
    Click **Create API key**. Give it a name, then either:

    * select an existing **Google Cloud project** from the dropdown, or
    * choose **Create API key in new project** for a clean project.

    Click **Create key**.
  </Step>

  <Step title="Copy the key">
    Click the key's name in the list to open its details, then use the **copy** icon to grab the full key. You can return to this page to copy it again later.
  </Step>

  <Step title="Set up billing for paid models">
    The Gemini API has a free tier, but the paid models — **especially Veo video** — require billing. From AI Studio, open **Billing** (or the linked Google Cloud project) and add a payment method. Confirm the key's project can access the Gemini 3.x text models, the Gemini 3.1 Flash Image model, and the Veo 3.1 video models.

    <Note>
      Google charges you directly for usage — this is separate from your MagicAds platform credits.
    </Note>
  </Step>
</Steps>

## Part 2 — Add the key in MagicAds

<Steps>
  <Step title="Open AI Settings">
    In your platform, go to **Admin → AI Settings**.
  </Step>

  <Step title="Open the Google vendor">
    Click the **Google** vendor card to open its configuration modal.
  </Step>

  <Step title="Paste the key">
    Paste your API key into the **API key** field. Leaving it blank keeps the existing key unchanged.
  </Step>

  <Step title="Enable the models you want">
    Toggle on the Gemini/Veo models you want to offer across the Copy, Image, and Video studios, and set each model's **credit cost**.
  </Step>

  <Step title="Save">
    Click **Save**. The Google card shows as connected and the enabled models appear in their studios.
  </Step>
</Steps>

<Note>
  There is no "test connection" button. To verify, enable a model and run a quick generation in the matching studio.
</Note>

## Part 3 — Supported Google models

### Copy Studio

| Model                 | Model ID                 | Tier     | Default credits | Best for                                                         |
| --------------------- | ------------------------ | -------- | --------------- | ---------------------------------------------------------------- |
| Gemini 3.1 Pro        | `gemini-3.1-pro-preview` | Premium  | 3               | Highest-quality Gemini; long-form, multilingual, complex briefs. |
| Gemini 3.5 Flash      | `gemini-3.5-flash`       | Standard | 1               | Near-Pro intelligence at Flash speed/cost; strong general pick.  |
| Gemini 3.1 Flash-Lite | `gemini-3.1-flash-lite`  | Fast     | 1               | Fastest, most budget-friendly; high-volume short copy.           |

### Image Studio

| Model                | Model ID                 | Default credits | Notes                                                                                                                                                      |
| -------------------- | ------------------------ | --------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Google Nano Banana 2 | `gemini-3.1-flash-image` | 2               | Best overall realism with character consistency, native 4K, strong editing and multilingual text. The default image model. Supports reference-image input. |

### Video Studio

| Model          | Model ID                        | Default credits | Notes                                                                                          |
| -------------- | ------------------------------- | --------------- | ---------------------------------------------------------------------------------------------- |
| Google Veo 3.1 | `veo-3.1-generate-preview`      | 12              | Film-grade 1080p with native synchronized audio/dialogue in one pass. The default video model. |
| Veo 3.1 Lite   | `veo-3.1-fast-generate-preview` | 6               | Faster, cheaper Veo tier for quick iterations; same engine, lower fidelity.                    |

### Also used by plugins

* **Fashion Studio** uses Gemini's Nano Banana 2 as its image engine and **Veo 3.1 Lite** for its Create Video tool.
* **Product Photoshoot** uses Gemini's Nano Banana 2 as its image engine.

Both read the same Gemini key, so configuring it here also unlocks those plugin features (subject to each plugin's own settings).

<Note>
  Credit costs shown are the seeded defaults. You can change any model's cost in the vendor modal. These are what your platform charges users in credits — not Google's prices.
</Note>

## Part 4 — Make Gemini a default (optional)

In the global section of **AI Settings**, Gemini is already the default for image and video out of the box (default image model = Google Nano Banana 2, default video model = Veo). You can also set the **default copy engine** to Gemini, toggle each studio on/off, and control free-tier access there.

## Troubleshooting

| Symptom                                            | Likely cause                                                    | Fix                                                                      |
| -------------------------------------------------- | --------------------------------------------------------------- | ------------------------------------------------------------------------ |
| Gemini/Veo models don't appear                     | Key missing or models disabled                                  | Add the key in the Google vendor modal and enable the models, then Save. |
| Image and copy work, but Veo video fails           | Billing not enabled for Veo                                     | Enable billing on the Google project; Veo requires it.                   |
| "The payload is invalid." on generate              | Saved key can't be decrypted (e.g. after an environment change) | Re-enter the Gemini key in AI Settings and Save to refresh it.           |
| Fashion Studio / Product Photoshoot can't generate | Gemini key not set                                              | Add the Gemini key in AI Settings (those plugins read the same key).     |
| Gemini API errors / 401 / 403                      | Invalid key or no model access                                  | Re-check the key and that its project can access the Gemini/Veo models.  |

<Note>
  Keys are stored encrypted and applied at runtime. No `.env` changes or commands are needed after saving.
</Note>