Omniroute
Never stop coding. Free AI gateway: one endpoint, 160+ providers (50+ free), connect Claude Code, Codex, Cursor, Cline & Copilot to FREE Claude/GPT/Gemini. RTK+Caveman stacked compression saves 15-95% tokens, smart auto-fallback, MCP/A2A, multimodal
What is Omniroute?
Omniroute is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to never stop coding. free ai gateway: one endpoint, 160+ providers (50+ free), connect claude code, codex, cursor, cline & copilot to free claude/gpt/gemini. rtk+caveman stacked compression saves 15-95%...
Never stop coding. Free AI gateway: one endpoint, 160+ providers (50+ free), connect Claude Code, Codex, Cursor, Cline & Copilot to FREE Claude/GPT/Gemini. RTK+Caveman stacked compression saves 15-95% tokens, smart auto-fallback, MCP/A2A, multimodal
This server falls under the Cloud Services category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Never stop coding. Free AI gateway: one endpoint, 160+ provi
Use Cases
Maintainer
Works with
Installation
NPM
npx -y omnirouteManual Installation
npx -y omnirouteConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Omniroute
OmniRoute is a free, self-hosted AI gateway that aggregates 177+ LLM providers — including 50+ free-tier and 11 permanently-free options — behind a single OpenAI-compatible endpoint, eliminating API-key exhaustion for developers who use tools like Claude Code, Cursor, Cline, Copilot, or Codex. It ships an MCP server with 37 tools across 3 transport protocols, 14 routing strategies (cost-optimized, weighted, context-relay, and more), stacked RTK+Caveman token compression saving 15–95% of tokens, automatic provider failover with circuit breakers, and a local dashboard for full observability. Developers use it to stay unblocked when rate limits hit and to slash inference costs across their entire toolchain.
Prerequisites
- Node.js 18+ (for npx install) or Docker
- No mandatory API keys — OmniRoute ships with 50+ free providers ready to use
- Optional: API keys for paid providers (Anthropic, OpenAI, xAI, DeepSeek, Groq, etc.) for higher rate limits
- An MCP-compatible client such as Claude Desktop, Claude Code, Cursor, or Cline
- Port 20128 available on localhost
Install and start OmniRoute
Install OmniRoute globally via npm and launch the gateway. It starts the API server and dashboard on port 20128 automatically.
npm install -g omniroute
omnirouteOr run with Docker (no Node.js required)
Use the official Docker image for a fully isolated deployment with persistent data.
docker run -d --name omniroute --restart unless-stopped \
-p 20128:20128 \
-v omniroute-data:/app/data \
diegosouzapw/omniroute:latestOpen the dashboard and add optional API keys
Navigate to http://localhost:20128 to access the OmniRoute dashboard. Free providers work immediately with no configuration. To add paid providers, enter your API keys in the Providers section for higher quotas.
# Open: http://localhost:20128Start the MCP server
OmniRoute includes a built-in MCP server exposing 37 tools. Launch it alongside the gateway to enable MCP client connections.
omniroute --mcpRegister OmniRoute as an MCP server via CLI
Use the claude mcp command to register OmniRoute's streaming HTTP endpoint with your Claude Code installation.
claude mcp add-server omniroute --type http \
--url http://localhost:20128/api/mcp/streamConfigure claude_desktop_config.json
For Claude Desktop, add OmniRoute to your MCP server config to route all LLM calls through the gateway with automatic fallback and token compression.
{
"mcpServers": {
"omniroute": {
"command": "npx",
"args": ["-y", "omniroute", "--mcp"],
"env": {
"PORT": "20128"
}
}
}
}Omniroute Examples
Client configuration
claude_desktop_config.json block for connecting Claude Desktop to a local OmniRoute MCP server.
{
"mcpServers": {
"omniroute": {
"command": "npx",
"args": ["-y", "omniroute", "--mcp"],
"env": {
"PORT": "20128",
"REQUIRE_API_KEY": "false"
}
}
}
}Prompts to try
Sample prompts once OmniRoute is running and connected as an MCP server.
- "Route this prompt to the cheapest free model available right now."
- "Show me my current token usage and savings from compression this session."
- "Switch to the cost-optimized routing strategy for the next 10 requests."
- "List all 50 free-forever providers currently online in OmniRoute."
- "Enable PII guardrails so sensitive data is never sent to external providers."Troubleshooting Omniroute
Gateway returns errors immediately after startup
Run `omniroute` without the --mcp flag first to verify the core server starts. Check that port 20128 is not already in use (`lsof -i :20128`). Set a different port with the PORT environment variable if needed.
MCP client cannot connect to OmniRoute
Ensure OmniRoute was started with the --mcp flag. Verify the URL in your MCP config is http://localhost:20128/api/mcp/stream (not just /api/mcp). Restart your MCP client after any config change.
All free providers are rate-limited simultaneously
OmniRoute's circuit breaker will pause exhausted providers and automatically fall back to the next available one. Add at least one paid provider API key (e.g., ANTHROPIC_API_KEY or OPENAI_API_KEY) in the dashboard to ensure continuous availability.
Frequently Asked Questions about Omniroute
What is Omniroute?
Omniroute is a Model Context Protocol (MCP) server that never stop coding. free ai gateway: one endpoint, 160+ providers (50+ free), connect claude code, codex, cursor, cline & copilot to free claude/gpt/gemini. rtk+caveman stacked compression saves 15-95% tokens, smart auto-fallback, mcp/a2a, multimodal It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Omniroute?
Install via npm with the command: npx -y omniroute. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).
Which AI clients work with Omniroute?
Omniroute works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Omniroute free to use?
Yes, Omniroute is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
Omniroute Alternatives — Similar Cloud Services Servers
Looking for alternatives to Omniroute? Here are other popular cloud services servers you can use with Claude, Cursor, and VS Code.
Open WebUI
★ 138.2kUser-friendly AI Interface (Supports Ollama, OpenAI API, ...)
Anything LLM
★ 60.4kThe all-in-one AI productivity accelerator. On device and privacy first with no annoying setup or configuration.
LocalAI
★ 46.4kLocalAI is the open-source AI engine. Run any model - LLMs, vision, voice, image, video - on any hardware. No GPU required.
Nacos
★ 33.0kan easy-to-use dynamic service discovery, configuration and service management platform for building AI cloud native applications.
Xiaozhi ESP32
★ 26.7k本项目为xiaozhi-esp32提供后端服务，帮助您快速搭建ESP32设备控制服务器。Backend service for xiaozhi-esp32, helps you quickly build an ESP32 device control server.
Gateway
★ 11.8kA blazing fast AI Gateway with integrated guardrails. Route to 1,600+ LLMs, 50+ AI Guardrails with 1 fast & friendly API.
Browse More Cloud Services MCP Servers
Explore all cloud services servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Omniroute in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Omniroute?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.