Home›Third-party Integrations

Configure DeepSeek API for Cherry Studio, Chatbox, and NextChat

Connect DeepSeek API to desktop chat clients and web interfaces for local document Q&A, knowledge base vector search, and web browsing workflows.

Maintained by DeepSeekDSH (independent site)Updated: 2026-09-13

On this page

OpenAI-compatible parameters

Base URL: https://api.deepseek.com or https://api.deepseek.com/v1
Chat model: deepseek-chat (DeepSeek-V3)
Reasoning model: deepseek-reasoner (DeepSeek-R1)
API key: Generated in your DeepSeek Platform console.

CLIENT 1

Cherry Studio setup (Desktop UI)

Cherry Studio is a multi-provider desktop client with built-in knowledge base support:

Open Cherry Studio and navigate to Settings ⚙️ → Model Providers.
Select DeepSeek from the provider catalog.
Enter your API Key (sk-...).
Click Check Connection to verify that deepseek-chat and deepseek-reasoner are active.

CLIENT 2

Chatbox setup (Cross-platform client)

Launch Chatbox and click Settings in the bottom left.
Under AI Model Provider, choose DeepSeek API (or OpenAI API Compatible).
Paste your API Key.
Set the API Host to https://api.deepseek.com if using custom mode.
Select deepseek-chat or deepseek-reasoner and save.

CLIENT 3

NextChat (ChatGPT-Next-Web) / LobeChat

For self-hosted web deployments using Docker:

# Docker run NextChat with DeepSeek
docker run -d -p 3000:3000 \
  -e BASE_URL=https://api.deepseek.com \
  -e OPENAI_API_KEY=sk-your_deepseek_api_key \
  -e CUSTOM_MODELS="-all,+deepseek-chat,+deepseek-reasoner" \
  yidadaa/chatgpt-next-web

Parameter tuning recommendations

Coding and math tasks

Set Temperature to 0.0 ~ 0.2 for high determinism and exact syntax output.

General chat and writing

Set Temperature to 0.6 ~ 0.7 for balanced creativity and factual accuracy.

Troubleshooting

401 Unauthorized

Verify that the key string starting with sk- has no extra spaces.

402 Insufficient Balance

Your API balance is exhausted. Top up your account in the DeepSeek Platform.