English | 简体中文

qwen2API

Self-hosted Qwen Web protocol gateway with OpenAI, Anthropic, and Gemini compatible APIs.

GitHub · Docker Hub · Telegram · 中文说明

qwen2API converts Qwen Web capabilities into common API protocols and provides a local WebUI for account management, downstream API keys, runtime settings, model tests, image tests, and video tests.

For most deployments, use the Docker Hub image directly. Keep data and logs beside your compose file; Docker will mount them into the container so upgrades do not wipe your accounts, keys, or logs.

mkdir qwen2api
cd qwen2api
mkdir -p data logs

Create a small .env:

HOST_PORT=7860
HOST_DATA_DIR=./data
HOST_LOGS_DIR=./logs
ADMIN_KEY=replace-with-your-own-strong-random-key

Create docker-compose.yml:

services:
  qwen2api:
    image: ${QWEN2API_IMAGE:-yujunzhixue/qwen2api:latest}
    container_name: qwen2api
    restart: unless-stopped
    init: true
    env_file:
      - .env
    ports:
      - "${HOST_PORT:-7860}:${PORT:-7860}"
    volumes:
      - ${HOST_DATA_DIR:-./data}:/app/data
      - ${HOST_LOGS_DIR:-./logs}:/app/logs
    shm_size: "512m"
    healthcheck:
      test: ["CMD-SHELL", "curl -fsS http://127.0.0.1:${PORT:-7860}/healthz || exit 1"]
      interval: 30s
      timeout: 10s
      start_period: 120s
      retries: 3

You do not need to set paths for accounts.json, api_keys.json, or other internal files. The image already uses /app/data and /app/logs; the volume mapping above decides where those files live on your host.

Start it:

docker compose pull
docker compose up -d
docker compose logs -f qwen2api

Open:

• WebUI: http://127.0.0.1:7860
• Health check: http://127.0.0.1:7860/healthz
• Keepalive probe: http://127.0.0.1:7860/keepalive

Use this path when you changed the source code and need to build your own image.

git clone https://github.com/YuJunZhiXue/qwen2API.git
cd qwen2API
cp .env.example .env
docker compose -f docker-compose.yml -f docker-compose.build.yml build
docker compose -f docker-compose.yml -f docker-compose.build.yml up -d

flowchart LR
  subgraph Clients["API Clients"]
    OpenAI["OpenAI SDK / Chat Completions"]
    Anthropic["Claude / Anthropic Messages"]
    Gemini["Gemini-compatible clients"]
    CLI["Claude Code / Codex / other CLI tools"]
  end

  subgraph App["qwen2API v2.0"]
    WebUI["React WebUI"]
    Router["Go HTTP Router"]
    Adapter["Protocol Adapters"]
    Tools["Tool-call / Context Pipeline"]
    Pool["Qwen Account Pool"]
    Store["JSON Stores / Data Files"]
  end

  subgraph Runtime["Runtime"]
    Docker["Docker image"]
    Data["./data volume"]
    Logs["./logs volume"]
  end

  Qwen["Qwen Web Upstream"]

  OpenAI --> Router
  Anthropic --> Router
  Gemini --> Router
  CLI --> Router
  WebUI --> Router
  Router --> Adapter
  Adapter --> Tools
  Tools --> Pool
  Pool --> Qwen
  Pool --> Store
  Store --> Data
  Router --> Logs
  Docker --> App

Do not commit real secrets. .env.example intentionally contains empty values and commented examples only.

• Go 1.26
• Node.js 20+
• npm
• Docker, only if you need container builds

go run start-all.go

cd backend
go run .

Verification:

cd backend
go test ./...
go build -trimpath -ldflags="-s -w" -o ..\bin\qwen2api-backend.exe .

cd frontend
npm ci
npm run dev

Production build:

cd frontend
npm run build

• Keep the Go backend as the v2.0 runtime source of truth.
• Keep Docker data paths container-internal as /app/data and /app/logs.
• Control host paths through compose volume mappings instead of hard-coded workspace paths.
• Do not commit data/, logs/, .env, real tokens, cookies, passwords, or downstream API keys.
• Update README and .env.example when adding user-visible configuration.

• Report bugs through GitHub Issues.
• Submit feature requests through GitHub Issues.
• Open focused pull requests through GitHub Pull Requests.
• Include practical verification steps when possible.

• go test ./... passes in backend.
• npm run build passes in frontend.
• Docker-related changes are reflected in Dockerfile, docker-compose.yml, and README when needed.
• No generated data, logs, local .env, Qwen token, cookie, password, or downstream API key is included.

Thanks to everyone who helps improve qwen2API.

This project is released under the GPL-3.0 License.

• This project is provided as an open-source self-hosted gateway.
• Review your local laws, platform rules, and upstream account policies before deployment.
• Do not publish or share real account tokens, cookies, passwords, or downstream API keys.
• If you find a security issue, please avoid public secret disclosure and report it through a private channel first.

• 特别鸣谢: LinuxDo