> For the complete documentation index, see [llms.txt](https://gpt-protocol.gitbook.io/gpt-protocol/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://gpt-protocol.gitbook.io/gpt-protocol/2.-products/2.1-gpt-assistant-your-ai-companion/2.1.2-getting-started.md).

# 2.1.2 Getting Started

GPT Assistant is an open-source platform utilizing AI and Blockchain technologies. To get started, follow the steps below.

<figure><img src="https://camo.githubusercontent.com/9ee14f138a49522772a66f29f928d00527335392ed115cf97afc6c329c3f88b6/68747470733a2f2f692e6962622e636f2f6e6663323352672f4750542d417373697374616e742e706e67" alt=""><figcaption><p>GPT Assistant 2.0</p></figcaption></figure>

**Initial Steps:**

1. **Clone the Repository:** Start by cloning the GPT Assistant repository from GitHub (<https://github.com/gptprotocol/gpt-assistant>). This can be done by using Git commands or directly downloading the repository as a ZIP file.
2. **Environment Setup:** After cloning, create a `.env.local` file in the root directory. This file will hold important configurations such as the MongoDB URL and the Hugging Face API token. Follow the current `.env` example file pattern.

**Database Configuration:**

1. **Setting Up MongoDB:** MongoDB is crucial for storing the chat history. You can set up a local MongoDB instance or use a MongoDB Atlas instance. For a local setup, using Docker to run a MongoDB container is an easy option.
2. **Configuring Database URL:** In the `.env.local` file, specify the MongoDB URL (`MONGODB_URL`). For a local MongoDB instance, the URL would typically be `mongodb://localhost:27017`.

**Installing Dependencies and Running the Application:**

1. **Install Dependencies:** Run `npm install` in the terminal to install all the necessary dependencies.
2. **Starting the Application:** After setting up the environment and database, start the application using `npm run dev`. This will launch the GPT Assistant interface.

**Additional Configurations:**

1. **API Keys for Web Search:** If you want to enable web search capabilities, add the appropriate API keys (like YDC\_API\_KEY, SERPER\_API\_KEY) to your `.env.local`.
2. **Custom Model Endpoints:** You can also configure custom model endpoints in the `.env.local` file. This allows for greater flexibility in terms of the models you can use with the Assistant.

This overview provides a simplified guide to getting started with the GPT Assistant. For complete details and step-by-step instructions, refer to the repository's README and documentation.