title	OmniBridge AI Proxy
emoji	⚡
colorFrom	indigo
colorTo	purple
sdk	docker
app_port	8000
pinned	false

OmniBridge

🇮🇷 برای مطالعه راهنمای فارسی اینجا کلیک کنید

A high-performance, self-hosted AI proxy server that translates Google Gemini, DeepSeek (V3 & R1 with automatic PoW solver), and ChatGPT web sessions into fully OpenAI-compatible and Anthropic-compatible API endpoints — complete with real-time reasoning_content (thinking) streams and a beautiful Persian admin dashboard.

📢 Official Telegram Channel: @Config_Vortex55

Features · Quick Start · API Reference · Admin Dashboard · Deploy

---

✨ Features

Feature	Details
🌐 Multi-Provider Engine	Seamlessly proxy Google Gemini, DeepSeek (V3 & R1), and ChatGPT
🧠 DeepSeek PoW Solver	Pure-Python 23-round Keccak-f[1600] solver handles DeepSeekHashV1 automatically
💡 Reasoning Stream	Real-time reasoning_content thinking tokens for DeepSeek-R1 and Gemini Extended
🤝 OpenAI Compatible	Drop-in replacement for POST /v1/chat/completions and GET /v1/models
🔶 Anthropic Compatible	Full POST /anthropic/v1/messages support
🌊 Real-time Streaming	Token-by-token SSE streaming preserving 100% of formatting, indentation, and spaces
🔒 TLS Fingerprinting	curl-cffi Chrome-120 impersonation — bypasses bot detection
🔄 Account Pooling	Round-robin load balancing + automatic cooldown failover across all accounts
🍪 Cookie Auto-Refresh	Background task validates & refreshes session cookies automatically
🗝️ API Key Auth	Generate/revoke sk-omni-... keys with full audit trail
📊 Admin Dashboard	Glassmorphic dark-mode Persian UI with live traffic charts and playground
🖼️ Multi-modal	Image input support via Gemini Vision models
🐋 Docker Ready	Multi-stage Dockerfile + compose + Render Blueprint

🏗️ Architecture

┌─────────────────────────────────────────────────────┐
│                    OmniBridge                       │
│                                                     │
│  ┌──────────┐    ┌──────────────────────────────┐   │
│  │ OpenAI   │    │         FastAPI App          │   │
│  │ /v1/...  │───▶│                              │   │
│  └──────────┘    │  ┌────────────────────────┐  │   │
│  ┌──────────┐    │  │    Gemini Engine       │  │   │
│  │Anthropic │───▶│  │  + Account Pool        │  │   │
│  │ /anth... │    │  │  + Cookie Auto-Refresh │  │   │
│  └──────────┘    │  └────────────────────────┘  │   │
│  ┌──────────┐    │         │                    │   │
│  │  Admin   │───▶│         ▼                    │   │
│  │  /admin  │    │  SQLite (aiosqlite) DB       │   │
│  └──────────┘    └──────────────────────────────┘   │
└─────────────────────────────────────────────────────┘

🚀 Quick Start

Local (Python 3.11+)

git clone https://github.com/exelite-dev/Free-Gemini-pro-API
cd Free-Gemini-pro-API

# Install dependencies
pip install -r requirements.txt

# Configure
cp .env.example .env
# Edit .env: set ADMIN_PASSWORD and ADMIN_SECRET_KEY

# Run
uvicorn app.main:app --host 0.0.0.0 --port 8000 --reload

Open http://localhost:8000/admin → login (admin / changeme) → add your Gemini account.

Docker Compose

docker compose up -d

Docker (single container)

docker build -t omnibridge .
docker run -d -p 8000:8000 \
  -e ADMIN_PASSWORD=mysecretpassword \
  -e ADMIN_SECRET_KEY=$(openssl rand -hex 32) \
  -v $(pwd)/data:/app/data \
  omnibridge

🎛️ Configuration

Environment Variables

Variable	Default	Description
ADMIN_USERNAME	admin	Admin dashboard username
ADMIN_PASSWORD	changeme	Admin dashboard password (change this!)
ADMIN_SECRET_KEY	supersecretkey	Session signing key (use a long random string)
GEMINI_ENABLED	true	Enable Gemini provider
DATABASE_PATH	./data/omnibridge.db	SQLite database path
PORT	8000	Server port
LOG_LEVEL	info	Logging verbosity

All settings can also be configured in config.yaml.

📡 API Reference

Authentication

All API endpoints require a Bearer token:

Authorization: Bearer sk-omni-<your-key>

Generate keys at /admin → API Keys tab.

List Models

GET /v1/models
Authorization: Bearer sk-omni-...

Returns all models for currently enabled providers.

Chat Completions (OpenAI format)

POST /v1/chat/completions
Content-Type: application/json
Authorization: Bearer sk-omni-...

{
  "model": "gemini-3.6-flash",
  "messages": [{"role": "user", "content": "Hello!"}],
  "stream": true
}

Available Gemini models:

Model ID	Description
gemini-3.6-flash	Fastest all-rounder (recommended default)
gemini-3.1-pro	Advanced reasoning, math & code
gemini-pro-extended	Extended thinking / deep reasoning
gemini-3.5-flash-lite	Ultra-fast, minimal latency
gemini-3-pro	Gemini 3 Pro
gemini-3-flash	Gemini 3 Flash
gemini-2.5-pro	Gemini 2.5 Pro
gemini-2.5-flash	Gemini 2.5 Flash
gemini-1.5-pro	Gemini 1.5 Pro
gemini-1.5-flash	Gemini 1.5 Flash

Messages (Anthropic format)

POST /anthropic/v1/messages
Content-Type: application/json
Authorization: Bearer sk-omni-...

{
  "model": "gemini-3.6-flash",
  "messages": [{"role": "user", "content": "Hello!"}],
  "max_tokens": 2048
}

cURL Example

curl http://localhost:8000/v1/chat/completions \
  -H "Authorization: Bearer sk-omni-xxxx" \
  -H "Content-Type: application/json" \
  -d '{"model":"gemini-3.6-flash","messages":[{"role":"user","content":"Say hello"}],"stream":false}'

🛠️ Admin Dashboard

Access at http://localhost:8000/admin

Tab	Description
Dashboard	Live request volume chart, success rate, latency, provider health
Accounts	Add/edit/delete Gemini accounts, test connections
Cookie Guide	Step-by-step cookie extraction guide with screenshots
API Keys	Generate sk-omni-... keys, revoke old keys
Integration	Copy-ready config for Cursor IDE, NextChat, Python scripts
Playground	Live streaming chat — test any model directly
Deploy Guide	Hosting instructions for Render, Hugging Face, VPS

🍪 Getting Gemini Credentials

1. Log into gemini.google.com with your Google account
2. Open DevTools (F12) → Application tab → Cookies → gemini.google.com
3. Copy the values of __Secure-1PSID and __Secure-1PSIDTS
4. Go to Admin Dashboard → Accounts → add them

Tip: The admin panel has a built-in visual guide with screenshots under the Cookie Guide tab.

🚢 One-Click Deploy

Render (Recommended — Free)

1. Fork this repository
2. Click the button above (or go to render.com → New → Web Service → connect your fork)
3. Set runtime to Docker
4. Add environment variables: ADMIN_PASSWORD, ADMIN_SECRET_KEY
5. Click Create Web Service — your proxy is live!

Railway

VPS / Self-Hosted

git clone https://github.com/exelite-dev/Free-Gemini-pro-API
cd Free-Gemini-pro-API
cp .env.example .env
nano .env  # set ADMIN_PASSWORD and ADMIN_SECRET_KEY
docker compose up -d --build

Hugging Face Spaces (Docker SDK)

1. Create a new Space → SDK: Docker
2. Push or upload the project files
3. In the Dockerfile, change EXPOSE 8000 to EXPOSE 7860 and update the CMD port accordingly
4. Set Space Secrets for ADMIN_PASSWORD and ADMIN_SECRET_KEY

📁 Project Structure

OmniBridge/
├── app/
│   ├── main.py              # FastAPI app factory + lifespan
│   ├── config.py            # Config (env + YAML, single source of truth)
│   ├── database.py          # Async SQLite (accounts, keys, metrics)
│   ├── models.py            # Pydantic schemas (OpenAI + Anthropic)
│   ├── auth.py              # API key + admin session auth
│   ├── providers/
│   │   ├── base.py          # Abstract provider + exceptions
│   │   ├── pool.py          # Round-robin account pool + failover
│   │   ├── registry.py      # Provider registry
│   │   └── gemini/
│   │       ├── engine.py    # curl-cffi TLS, cookie auth, streaming
│   │       └── cookie_refresh.py  # Background auto-refresh task
│   ├── routes/
│   │   ├── openai.py        # /v1/chat/completions, /v1/models
│   │   ├── anthropic.py     # /anthropic/v1/messages
│   │   ├── google_proxy.py  # /google/* pass-through proxy
│   │   └── admin.py         # Admin REST API + SPA serving
│   └── admin/
│       ├── static/css/      # Glassmorphic design system
│       └── templates/       # Admin SPA (Jinja2 + vanilla JS)
├── tests/
│   └── test_app.py          # Unit & integration tests
├── config.yaml              # Default provider/model config
├── .env.example             # Environment variable template
├── requirements.txt
├── Dockerfile               # Multi-stage build
├── docker-compose.yml
├── render.yaml              # Render Blueprint
├── README.md
└── README.fa.md             # Persian README

🔒 Security Notes

• Change ADMIN_PASSWORD and ADMIN_SECRET_KEY before any public deployment
• Never commit .env to version control (it's in .gitignore)
• In production, run behind a reverse proxy (nginx / Caddy) with TLS termination
• The default sk-test key is for local development only — revoke or replace it in production
• Gemini cookies are stored encrypted-equivalent in SQLite; rotate them via the admin panel periodically

🧪 Running Tests

pip install pytest pytest-anyio
python -m pytest tests/ -v

📄 License

MIT License — see LICENSE for details.

📢 Community

• Telegram: @Config_Vortex55 — updates, tips, and support