Using Vercel AI Gateway in LobeHub

Vercel AI Gateway is a unified API that provides access to over 100 AI models through a single endpoint. It offers features such as budget management, usage monitoring, load balancing, and fallback handling.

This guide will walk you through how to use Vercel AI Gateway within LobeHub.

<Steps> ### Step 1: Create an API Key in Vercel AI Gateway

Go to the Vercel Dashboard
Click on the AI Gateway tab on the left sidebar
Select API Keys from the sidebar
Click Create Key, then confirm by clicking Create Key in the dialog

Step 2: Configure Vercel AI Gateway in LobeHub

Navigate to the Settings page in LobeHub
Under AI Service Provider, find the Vercel AI Gateway section
Enter the API key you obtained
Choose a model from Vercel AI Gateway and start chatting with the AI assistant

<Callout type={'warning'}> You may incur charges from the API service provider during usage. Please refer to the Vercel AI Gateway pricing policy for details. </Callout> </Steps>

That's it! You can now start chatting in LobeHub using models provided by Vercel AI Gateway.

Model Selection

Vercel AI Gateway supports a variety of model providers, including:

OpenAI: openai/gpt-4o, openai/gpt-4o-mini, openai/o1, and more
Anthropic: anthropic/claude-3-5-sonnet, anthropic/claude-3-opus, and more
Google: google/gemini-2.5-pro, google/gemini-2.0-flash, and more
DeepSeek: deepseek/deepseek-chat, deepseek/deepseek-reasoner, and more
And many others...

To view the full list of supported models, visit the Vercel AI Gateway Models page.

API Configuration

Vercel AI Gateway uses an OpenAI-compatible API format. The base URL is:

https://ai-gateway.vercel.sh/v1

You can use any OpenAI-compatible client with this endpoint and your API key.