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

# Vercel AI Gateway

> Connect Vercel AI Gateway and choose an AI model in Enconvo.

Use **Vercel AI Gateway** as an AI Model provider in Enconvo. The available model list comes from the installed provider and your account access; select a model in the app rather than relying on a fixed list in this guide.

## Connect

1. Open **Settings → Global Providers Settings → AI Model**.
2. Select **Vercel AI Gateway**, or use **Add New Provider** to create another profile.
3. Configure the provider’s **Authentication** using your own account or API key, following the available method on that page.
4. Validate the credential if a validation action is offered.
5. Choose the model or deployment, then select this provider in your Agent or conversation and send a short text request.

## Provider options

Use the gateway’s credential and select the routed model. Model capabilities, account access, and billing follow the selected route. Google Search Tool and URL Context Tool are available only on compatible models.

The provider definition includes: **Model Name**, **Temperature**, **Google Search Tool**, **URL Context Tool**. Some controls appear only for models or authentication methods that support them.

## Use it in a conversation

Select the model from the conversation’s model menu. Review reasoning and tool support for the selected model before running an Agent task. A text-only request is a useful connection check before adding images, documents, or tools.

Provider profiles share credentials across their supported categories. See [Credentials and Provider Profiles](/configuration/credentials) before renaming, deleting, or changing a shared key.

## Troubleshooting

| Symptom                     | Check                                                                     |
| --------------------------- | ------------------------------------------------------------------------- |
| Authentication failure      | Correct profile, authentication method, and account access                |
| Model unavailable           | Current model picker and access granted to your account                   |
| Tool or image request fails | Selected model capabilities and enabled Agent tools                       |
| Rate or usage limit         | Provider account usage, or Cloud usage if using an Enconvo Cloud provider |
| Slow response               | Input size, selected model, reasoning setting, and provider availability  |

See the [provider capability catalog](/providers/catalog) for other categories and [Model Selection](/ai/model-selection-guide) for choosing a model by task.