Cherry Studio is a stunning, cross-platform (Mac/Win/Linux) AI client from China. It supports managing multiple models in a sleek interface, making it perfect for power users of DeepSeek V3.
Why Cherry Studio?
- Beautiful UI: Modern, clean design with multiple themes (including cherry blossom pink).
- Model Arena: Compare answers from DeepSeek V3 vs Claude vs GPT-4 side-by-side.
- Local RAG: Drag and drop documents to chat with them immediately.
Configuration Steps
Cherry Studio has built-in presets for SiliconFlow (a high-performance provider for DeepSeek).
Method 1: Using SiliconFlow (Recommended)
- Get Key: Sign up at SiliconFlow and create an API Key.
- Open Settings: Click the gear icon -> Model Services.
- Select Provider: Enable SiliconFlow from the list.
- Enter Key: Paste your
sk-xxxxkey. - Add Model: Click "Add Model" and enter:
deepseek-ai/DeepSeek-V3(Chat)deepseek-ai/DeepSeek-R1(Reasoning)
Method 2: Official DeepSeek API
- In Model Services, select DeepSeek (or add "Custom OpenAI" if missing).
- API Key: Enter your official DeepSeek key.
- Base URL:
https://api.deepseek.com - Model ID: Add
deepseek-chatordeepseek-reasoner.
Advanced: View Thinking Process
For DeepSeek R1 (Reasoning model), enable "Show Thought Process" in settings. This allows you to see the <think> chain of thought before the final answer, which is fascinating to watch!
FAQ
Q: Can't find SiliconFlow models?
A: You may need to manually click "Check Model List" or type the model ID deepseek-ai/DeepSeek-V3 directly.
Q: Is it safe? A: Cherry Studio stores keys locally. It's a client-side app, so your data goes directly to the API provider.
