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

# Providers and models

> Bring your own API keys and select the model that works for your task.

AscentialxAI works with your own API keys from supported AI providers. You manage everything from the **Settings** > **API Providers** panel.

## Supported providers

* **OpenAI**
* **OpenRouter**
* **Straico**
* **Google** (Gemini models)

## Add a provider key

1. Open **Settings** > **API Providers**.
2. Select a provider from the list.
3. Enter your API key. You can find or create keys from the provider's API dashboard — links are available next to each provider field.
4. Click **Save**.

After saving, the extension syncs your available models automatically.

## Delete a provider key

Open **Settings** > **API Providers**, select the provider, and choose **Delete** to remove the saved key.

## Sync models

After adding or updating a key, models sync automatically. You can also trigger a manual refresh from the model selector.

## Select a model

The model selector is available in the chat composer toolbar and the attachment menu.

* **Search** — Type to filter models by name.
* **Filter by provider** — Narrow the list to a specific provider.
* **Favorites** — Star models for quick access.
* **Refresh** — Re-sync the model list if you've added new keys.

## No model configured

If you haven't connected a provider or no models are available, the composer shows a prompt to configure one. Open **Settings** > **API Providers** to get started.

## Vision support

Some models support vision — they can analyze images, screenshots, and page content. If you plan to use image attachments or `/vision`, make sure your selected provider and model support vision inputs.