How to set up LobeChat with Ozeki AI Gateway

This guide demonstrates how to run LobeChat using Docker and configure it to use Ozeki AI Gateway as its LLM provider. LobeChat supports custom OpenAI-compatible endpoints through its language model settings, and with client-side fetching enabled it can communicate directly with your gateway from the browser.

What is LobeChat?

LobeChat is an open-source, modern AI chat framework that provides a polished web-based interface for interacting with large language models. It supports a wide range of providers through OpenAI-compatible APIs and can be self-hosted using Docker with a single command.

Steps to follow

We assume Ozeki AI Gateway is already installed on your system. You can install it on Linux, Windows or Mac. You will also need Docker Desktop installed and running on your Windows machine.

Run LobeChat using Docker
Configure Ozeki AI Gateway in LobeChat
Send a test prompt

Docker run command

docker run -p 3210:3210 lobehub/lobe-chat

How to run and set up LobeChat with Ozeki AI Gateway video

The following video shows how to run LobeChat with Docker and configure it to use Ozeki AI Gateway step-by-step.

Step 1 - Run LobeChat using Docker

Open a terminal window and run the following Docker command to pull and start the LobeChat container. The -p 3210:3210 flag maps the container port to your local machine, making the interface accessible at http://localhost:3210 (Figure 1).

docker run -p 3210:3210 lobehub/lobe-chat

Wait for Docker to pull the image and start the container. Once the terminal shows that the server is ready, click the url in the terminal or open your browser and navigate to http://localhost:3210 (Figure 2).

Figure 2 - Wait for LobeChat to start and open the URL in your browser

The LobeChat homepage will appear in your browser. The interface is now running and ready to be configured with Ozeki AI Gateway as a provider (Figure 3).

Step 2 - Configure Ozeki AI Gateway in LobeChat settings

Open the settings by clicking the LobeChat logo in the top-left corner of the interface and selecting Settings (Figure 4).

Open LobeChat settings — Figure 4 - Open the LobeChat settings

Navigate to the Language Model section where you can configure AI provider endpoints. Locate the OpenAI provider section and enter your Ozeki AI Gateway connection details. Set the API URL to your gateway endpoint and paste your API key into the API key field (Figure 5).

http://localhost/v1

Enter Ozeki AI Gateway details in language model settings — Figure 5 - Enter the Ozeki AI Gateway details in the language model settings

Enable the Use Client-Side Fetching Mode toggle: this is required when LobeChat is running locally and needs to communicate directly with your gateway from the browser rather than routing through the container server. Once enabled, click the Get Model List button to fetch the list of models available from your gateway (Figure 6).

Enable client-side fetching and refresh model list — Figure 6 - Enable client-side fetching and refresh the model list

The models available through your Ozeki AI Gateway will now appear in the model list. Use the Connectivity Check tool to test the connection status (Figure 7).

Select model and verify connection — Figure 7 - Select a model and verify the connection

Step 3 - Send a test prompt

Start a new chat session in LobeChat and select the model provided by your Ozeki AI Gateway connection from the model selector at the top of the chat window (Figure 8).

Start new chat and select the configured model — Figure 8 - Start a new chat and select the configured model

Type a test message in the chat input and press the Send button or Enter to send it (Figure 9).

Send the LLM a test prompt — Figure 9 - Send a test prompt to the LLM

A successful response confirms that LobeChat is correctly connected to Ozeki AI Gateway and that all requests are being routed through your gateway infrastructure (Figure 10).

LLM response — Figure 10 - AI response received from Ozeki AI Gateway

Conclusion

You have successfully deployed LobeChat using Docker and configured it to use Ozeki AI Gateway as its LLM provider. With client-side fetching enabled, the browser communicates directly with your gateway, and all model requests are routed through your infrastructure for centralized access control and monitoring.