ds2api
SHA-256A Go-based middleware that adapts DeepSeek's web conversation engine to OpenAI, Claude, and Gemini API formats. It supports high-concurrency account pooling, PoW computation, and includes an admin WebUI.
Smart Download
Download Download Version
v4.6.1 · 39.7 MB
Convert DeepSeek Web into OpenAI/Claude/Gemini compatible APIs with a lightweight Go middleware.
Core Features
- Multi-protocol support: OpenAI, Claude, Gemini, and Ollama APIs from one endpoint
- Automatic account rotation and token refresh for better rate limits
- Built-in PoW calculation and concurrency queue control for stability
- Admin WebUI for real-time monitoring and runtime configuration
- Deployable via binary, Docker, or Vercel — works with any OpenAI-compatible client
What It Can't Do
- •For learning and research only — not authorized for commercial or ToS-violating use
- •Using your own DeepSeek account risks suspension; use throwaway accounts
- •Some upstream features (file uploads, web search) may not be fully supported
- •Upstream DeepSeek changes may break functionality; keep ds2api updated
Use Cases
- Developers who want to use DeepSeek models locally without OpenAI API costs, via standard SDKs
- Self-hosted AI gateway to manage multiple DeepSeek accounts and provide unified API for teams
Detailed Introduction
DS2API is a technical reference implementation in Go that acts as a middleware layer between various AI client SDKs (OpenAI, Claude, Gemini, Ollama) and DeepSeek's web conversation engine. It handles protocol adaptation, session management, account pooling, and queue control for high-concurrency scenarios. Unlike traditional API proxies (e.g., One API), DS2API directly interfaces with DeepSeek's web endpoints, supports PoW computation natively in Go, and provides a built-in admin WebUI for runtime configuration and account management. It is designed for learning, research, and personal experimentation, not for production commercial use.
Tags
Getting Started
Download installer
Click the button above to download the installer for your system
Install the software
Double-click the downloaded installer and follow the prompts
Download the archive for your OS from GitHub Releases and extract it
Rename config.example.json to config.json and fill in your DeepSeek credentials
Run ./ds2api; the console will show the listening port (default 5001)
- Download the archive for your OS from GitHub Releases and extract it
- Rename config.example.json to config.json and fill in your DeepSeek credentials
- Run ./ds2api; the console will show the listening port (default 5001)
Latest Release Notes
修复安全性漏洞 https://github.com/CJackHwang/ds2api/security/advisories/GHSA-rf34-c5jc-4ffw
工具解析稳定性补丁
## What's Changed
* [codex] fix security advisory and toolcall parsing issues by @CJackHwang in https://github.com/CJackHwang/ds2api/pull/476
* Fix tool detection when unclosed backtick precedes tool call by @CJackHwang in https://github.com/CJackHwang/ds2api/pull/480
* [codex] fix WebUI static root path guard by @CJackHwang in https://github.com/CJackHwang/ds2api/pull/481
**Full Changelog**: https://github.com/CJackHwang/ds2api/compare/v4.6.0...v4.6.1
SHA-256 checksum verified
Checksum extracted from GitHub official Release page
SHA256 Checksum
a73d983b72b8f04d59a8f50f06412e75fdbb7ac757f7b9171c11328b5e1d6166This checksum is extracted from the GitHub Release page. Verify file integrity after download.
All SHA-256 checksums on this platform are extracted from the project's official GitHub Release page, without any modification. You can independently verify them on the GitHub Releases page.
Open Source Transparency
View GitHub SourceUninstall Info
Delete the binary and configuration files to fully uninstall. For Docker deployments, run 'docker-compose down -v' to remove containers and networks.
No Extra Dependencies
Ready to use after download. No additional runtime required.
Similar Projects
LocalAI
LocalAI is the open-source AI engine to run any model - LLMs, vision, voice, image, video - on any hardware. No GPU required. Drop-in API compatibility with OpenAI, Anthropic, and ElevenLabs.
ollama
Ollama lets you download, run, and manage large language models locally. One command, multiple platforms, endless possibilities.
llama.cpp
High-performance LLM inference engine in C/C++ with minimal dependencies, supporting quantized models (1.5–8 bit) and diverse hardware (Apple Silicon, CUDA, Vulkan, etc.).