# How to Install Lobe Chat - Multi LLM Frontend Framework

### Overview

* Most developers start with the `OpenAI API` to develop using **LLM**. The **OpenAI API** operates separately from `ChatGPT` and allows the use of the latest model, `GPT-4 Turbo` (equivalent to `gpt-4-turbo-2024-04-09` as of December 2023).
    
* Developers using the **OpenAI API** often feel the need for a frontend similar to **ChatGPT**. This article introduces `Lobe Chat`, an open-source frontend that allows the **OpenAI API** to be used like **ChatGPT**. In my case, I install it locally for convenient, constant use.
    
* The advantage of **Lobe Chat** is that it provides an integrated frontend for almost all models worldwide. Just by entering the **API Key**, you can conveniently use it with a unified **UI**, which is highly recommended.
    

### Lobe Chat Installation

* **Lobe Chat** is very conveniently installed as it provides a Docker image. It does not require a separate database, and the memory usage during service operation is around **150MB**.
    

```bash
$ mkdir lobe-chat
$ cd lobe-chat

# Write Docker Compose file
$ nano docker-compose.yml
services:
  lobe-chat:
    image: lobehub/lobe-chat:latest
    container_name: lobe-chat
    restart: always
    ports:
      - '3210:3210'
    environment:
      OPENAI_API_KEY: {your-openai-api-key}
      ACCESS_CODE: {your-password}

# Install and run Lobe Chat service
$ docker-compose up -d
```

### Running Lobe Chat
  * You can access Lobe Chat by navigating to `http://localhost:3210` in a web browser.

### Using Amazon Bedrock Claude 3.7 Sonnet
  * You can add `Amazon Bedrock Claude 3.7 Sonnet` to the list of available models as follows:

```bash
Lobe Chat
→ [Settings]
→ [Language Model]
→ [Amazon Bedrock]
→ AWS Access Key Id: ***** (Enter)
→ AWS Secret Access Key: ***** (Enter)
→ AWS Region: Select [us-west-2]
→ Use Client-Side Fetching Mode: Select [On]
→ Model List: us.anthropic.claude-3-7-sonnet-20250219-v1:0 (Enter)
```

### References
* [GitHub - Lobe Chat](https://github.com/lobehub/lobe-chat)