A proxy that lets Claude Code and Agent SDK use domestic Chinese LLMs (DeepSeek, Qwen, GLM, MiniMax, Kimi) as backend
deepseek-proxy is published on npm as @sunflower0305/deepseek-proxy. It is a lightweight Express proxy that lets Claude Code or the DeepSeek Agent SDK talk to domestic Chinese LLM providers through Anthropic-compatible /v1/messages endpoints.
It currently supports qwen, deepseek, glm, minimax, and kimi.
Run without installing:
npx @sunflower0305/deepseek-proxy
Or install globally:
npm install -g @sunflower0305/deepseek-proxy
deepseek-proxy
The proxy reads configuration from environment variables. You can export them in your shell or create a .env file in the directory where you run deepseek-proxy.
Example .env:
PROVIDER=qwen
PROXY_PORT=8080
QWEN_API_KEY=your-qwen-api-key
QWEN_MODEL=qwen-plus
Available variables:
| Variable | Purpose |
|---|---|
PROVIDER | Active provider. Defaults to qwen. |
PROXY_PORT | Local server port. Defaults to 8080. |
QWEN_API_KEY | API key for Qwen. |
DEEPSEEK_API_KEY | API key for DeepSeek. |
GLM_API_KEY | API key for GLM. |
MINIMAX_API_KEY | API key for MiniMax. |
KIMI_API_KEY | API key for Kimi. |
QWEN_ANTHROPIC_BASE_URL, DEEPSEEK_ANTHROPIC_BASE_URL, GLM_ANTHROPIC_BASE_URL, MINIMAX_ANTHROPIC_BASE_URL, KIMI_ANTHROPIC_BASE_URL | Override the upstream Anthropic-compatible base URL for a provider. |
QWEN_MODEL, DEEPSEEK_MODEL, GLM_MODEL, MINIMAX_MODEL, KIMI_MODEL | Override the default upstream model for a provider. |
You can use the bundled example as a starting point:
cp node_modules/@sunflower0305/deepseek-proxy/.env.example .env
If you installed globally, create .env manually or export the variables in your shell before starting the proxy.
deepseek-proxy
When the server starts, it listens on http://localhost:8080 by default.
Point Claude Code or the DeepSeek Agent SDK at the proxy:
export ANTHROPIC_BASE_URL=http://localhost:80
Open-source AI coding agent for the terminal. Claude Code-grade accuracy with smart model routing — uses the right AI model for each task, cutting costs 10x. Supports Claude, GPT, Gemini, DeepSeek & more.
A general-purpose Python framework for building LLM agents and multi-agent systems. "Four lines of code, an agent with memory."
Ultimate LLM API Integration Cookbook 2026 for Cursor & AI Agents
Ultimate Multi-Agent OS for Autonomous AI NPCs 2026
PrivateAgent is an open-source Android automation agent built with Flutter. It utilizes the DeepSeek API and native Android Accessibility Services to interpret screen layouts and execute multi-step tasks across any installed application via natural language commands.
把一队分工 Agent 织成一条写小说的流水线,做成桌面客户端;写作指纹让它越写越像你(BYO DeepSeek key,纯本地)。
Workflows from the Neura Market marketplace related to this DeepSeek resource