OpenClaw Online Platform Deployment

1. Registering on an Online Model Platform

OpenClaw supports many online large language model platforms. The registration method varies depending on the model. Here, we take the Deepseek model as an example. Open your browser and visit DeepSeek Open Platform, register and log in to your account, and follow the instructions to authorize and obtain an API Key. Models require payment; please recharge before use.

https://platform.deepseek.com/sign_in

Click 【API keys】-> 【Create new API Key】.

Enter a name, and then click 【Create API key】.

After creation, you must copy the API key. It's best to save it in a notepad, as you'll need it when filling in the API key later.

Click "Top up". If you haven't completed real-name authentication yet, you need to authenticate first before topping up.

Click [Docs] to navigate to the DeepSeek API documentation.

2. Configuring OpenClaw

Open your terminal and execute the OpenClaw configuration command:

xxxxxxxxxx
openclaw onboard  --install-daemon

You must select [Yes] to proceed with the configuration process. Press [Enter] to proceed to the next step.

Select [QuickStart] to begin the quick setup.

On the model selection screen, select the specific model platform you wish to use. This step is mandatory; otherwise, you will not be able to interact with the system properly. Taking the DeepSeek model as an example, you need to select [Custom Provider].

Set the API Base URL to:

xxxxxxxxxx
https://api.deepseek.com

Select [Paste API Key now] and enter the API Key you generated on DeepSeek.

Select [OpenAI-compatible].

For the [Model ID], enter:

 OpenClaw  Installation
OpenClaw Local Model Deployment