FreeDeepseekAPI EN

FreeDeepseekAPI-EN
🌍 Global English Version: This is the English-translated and globally adapted version of the proxy. Core logic belongs to the original author (ForgetMeAI), maintained and translated for the global open-source community by Atharvotech.

FreeDeepseekAPI starts a local API server/wrapper for DeepSeek Web Chat (chat.deepseek.com) and lets you seamlessly connect DeepSeek Web to Open WebUI, LiteLLM, Hermes, Claude Code, OpenAI SDK-style clients, and other OpenAI-compatible LLM tools.

The project works through your regular logged-in DeepSeek account in a dedicated Chrome profile. The local localhost server accepts API requests and then talks to DeepSeek Web on its own via the saved browser session.
⚠️ This is an experimental web-chat proxy for local LLM integration. DeepSeek can change its internal Web API without warning. For production use, the official paid DeepSeek API is more reliable.

Table of Contents

- What this gives you
- Features
- Quick start
- Windows launch
- Linux / Chromium launch
- VPS / headless launch
- Rootless Podman
- Diagnostics / doctor
- Session reuse and chat reset
- Multi-account pool
- Console auth ideas
- Verifying it works
- Usage examples
  - Chat Completions
  - Reasoning
  - Web search
  - Streaming
  - Anthropic Messages API
  - OpenAI Responses API
  - Tool calling
- Models
- Endpoints
- Open WebUI
- Update login
- Project status

✨ What this gives you

- Use DeepSeek Web as a local API endpoint.
- Connect DeepSeek to Open WebUI and other OpenAI-compatible clients.
- Get regular JSON responses or streaming SSE.
- Use reasoning models with separate reasoning_content.
- Work with the Anthropic Messages API shim for Claude Code / Anthropic SDK.
- Use the OpenAI Responses API shim for new OpenAI/Codex-style clients.
- Keep…