DeepSeek, a powerful AI language model, offers a compelling alternative to standard web interfaces. While the official DeepSeek website provides a direct way to interact with the model, configuring it through an API unlocks a new level of flexibility and performance. This article provides a step-by-step guide to setting up DeepSeek using its API with Cherry Studio and Chatbox, addressing common questions and considerations.
While the web interface is convenient, using the DeepSeek API offers several advantages:
This guide focuses on using the DeepSeek API with Cherry Studio (for desktop) and Chatbox (for mobile and desktop), offering a smoother, more reliable, and customizable experience.
Cherry Studio is a robust desktop application that pairs nicely with the DeepSeekAPI. Here’s how to set it up:
Visit the Cherry Studio download page and select the appropriate version for your operating system. Once downloaded, install the application.
Open Cherry Studio and click the settings icon, located at the bottom-left of the interface.
Navigate to the "Model Services" section and choose "DeepSeek" from the available options. Enter your DeepSeek API key (refer to the official documentation for how to obtain an API Key).
Click "Check" to verify the connection. Once the API is confirmed to be working, activate it using the toggle switch in the upper-right corner. You can select a model to test.
You can now interact with DeepSeek-R1 in Cherry Studio. The platform and model can be selected at the top of the interface.
Go to Settings -> Model Services -> + Add. Now refer the the following list.
To integrate platforms like Silicon Flow, Alibaba Cloud, Baidu AI Cloud, or ByteDance Volcano Engine, use the following parameters:
| Parameter | Silicon Flow | Alibaba Cloud BaiLian | Baidu AI Cloud/Baidu Qianfan | ByteDance Volcano Engine |
|---|---|---|---|---|
| Acquisition | Instructions | Instructions | Instructions | Instructions |
| API Domain | https://api.siliconflow.cn |
https://dashscope.aliyuncs.com/compatible-mode/v1/ |
https://qianfan.baidubce.com/v2/ |
https://ark.cn-beijing.volces.com/api/v3/ |
| Model - Chat | deepseek-ai/DeepSeek-V3 |
deepseek-v3 |
deepseek-v3 |
Visit Inference Endpoint to obtain |
| Model - Inference | deepseek-ai/DeepSeek-R1 |
deepseek-r1 |
deepseek-r1 |
Visit Inference Endpoint to obtain |
Chatbox is an excellent option for both desktop and mobile users. Here's how to configure it with the DeepSeek API:
Download Chatbox from the official website or your device's app store (iOS, Android). Install the application on your device.
Open Chatbox, and select "Use your own API Key or local model."
In the settings, choose "DeepSeek API," then enter your DeepSeek API key and save.
You can now begin using DeepSeek-R1 within Chatbox.
You can add custom providers like Silicon Flow, Alibaba Cloud, Baidu AI Cloud, or ByteDance Volcano Engine using the same parameters listed in the Cherry Studio section.
Access "Add Custom Provider" through either:
Q: How do I add more models using a custom provider?
Q: How do I adjust the 'temperature' parameter?
Refer to the DeepSeek official documentation for recommended settings based on the use case:
Q: How do I switch platforms in Chatbox?
Q: How do I back up my data?
By configuring the DeepSeek API with Cherry Studio or Chatbox, you gain more control, flexibility, and reliability compared to relying solely on the web interface. These platforms allow you to tailor your AI interactions, switch between providers, and manage your data effectively. Experiment with custom providers and adjust parameters to discover the optimal setup for your AI needs.
