MCP Server

Active

MIT

GoModel MCP Server

io.github.ENTERPILOT/gomodel

Self-hosted AI gateway aggregating MCP servers and LLM providers behind one authenticated HTTP endpoint with caching, cost tracking, and observability.

Repository Registry

What is the GoModel MCP server?

GoModel is a self-hosted AI gateway that aggregates upstream MCP servers and multiple LLM providers (OpenAI, Anthropic, Cohere, Google Gemini, etc.) behind a single authenticated HTTP endpoint. It provides OpenAI-compatible and Anthropic-compatible APIs, enabling you to manage API keys, track costs, cache responses, and route requests across providers from one central point.

GoModel acts as a unified gateway for AI services, letting you aggregate MCP servers and LLM providers behind one endpoint with built-in authentication, caching, cost tracking, and failover. Use it to reduce API costs through response caching, monitor spending per user or team, enforce budgets and rate limits, and automatically route requests to backup providers if one fails.

How to install GoModel

Copy-paste configuration for popular MCP clients.

transport: stdio

Config generated by PluginBench — verify against the source before use.

Environment / auth

GOMODEL_MASTER_KEY
secret
Gateway API key clients authenticate with; unset runs the gateway in unsafe (no-auth) mode.
MCP_SERVERS
JSON object of upstream MCP servers to aggregate, e.g. {"github":{"url":"https://api.githubcopilot.com/mcp/","headers":{"Authorization":"Bearer ${GITHUB_PAT}"}}}. Servers can also be declared in config.yaml or the admin dashboard.

Claude Desktop

~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "gomodel": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "docker.io/enterpilot/gomodel:0.1.77"
      ],
      "env": {
        "GOMODEL_MASTER_KEY": "<YOUR_GOMODEL_MASTER_KEY>",
        "MCP_SERVERS": "<YOUR_MCP_SERVERS>"
      }
    }
  }
}

Cursor

~/.cursor/mcp.json

{
  "mcpServers": {
    "gomodel": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "docker.io/enterpilot/gomodel:0.1.77"
      ],
      "env": {
        "GOMODEL_MASTER_KEY": "<YOUR_GOMODEL_MASTER_KEY>",
        "MCP_SERVERS": "<YOUR_MCP_SERVERS>"
      }
    }
  }
}

Windsurf

~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "gomodel": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "docker.io/enterpilot/gomodel:0.1.77"
      ],
      "env": {
        "GOMODEL_MASTER_KEY": "<YOUR_GOMODEL_MASTER_KEY>",
        "MCP_SERVERS": "<YOUR_MCP_SERVERS>"
      }
    }
  }
}

VS Code

.vscode/mcp.json

{
  "servers": {
    "gomodel": {
      "type": "stdio",
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "docker.io/enterpilot/gomodel:0.1.77"
      ],
      "env": {
        "GOMODEL_MASTER_KEY": "<YOUR_GOMODEL_MASTER_KEY>",
        "MCP_SERVERS": "<YOUR_MCP_SERVERS>"
      }
    }
  }
}

Claude Code

claude mcp add gomodel --env GOMODEL_MASTER_KEY=<YOUR_GOMODEL_MASTER_KEY> --env MCP_SERVERS=<YOUR_MCP_SERVERS> -- docker run -i --rm docker.io/enterpilot/gomodel:0.1.77

Tools & capabilities

Tools this server exposes to the agent.

MCP Gateway — Aggregate upstream MCP servers behind one authenticated HTTP endpoint
OpenAI-compatible API — Expose OpenAI SDK-compatible endpoints at /v1
Anthropic-compatible API — Expose Anthropic SDK-compatible endpoints at /v1/messages
Passthrough API — Provider-native APIs under /p/{provider}/... with GoModel auth and tracking
Response Caching — Exact and semantic response caching to reduce repeated request costs
Cost Tracking — Per-request cost estimates, usage analytics, and spending breakdowns
Budgets — Hard spend limits per user, team, or API key
Rate Limiting — Request, token, and concurrency caps per user path, provider, or model
Virtual Models — Aliases and load balancing (round-robin or cost-based) behind stable model names
Failover & Retries — Automatic rerouting to backup providers with circuit breakers
Labelling — Tag requests from HTTP headers or API keys and break down usage by label
User Paths — Hierarchical scoping of keys, model access, budgets, usage, and audit logs
Guardrails — Request and response policies enforced at the gateway
Provider Key Rotation — Round-robin over multiple API keys to lift per-key rate limits
Observability — Prometheus metrics, audit logs, and live request streaming in dashboard

Use cases

Aggregate multiple MCP servers and LLM providers into one authenticated endpoint for your AI applications
Reduce API costs by caching responses and intelligently routing requests across providers
Monitor and control spending with per-user budgets, cost tracking, and usage analytics
Implement automatic failover and load balancing across multiple LLM providers to ensure reliability
Set rate limits and enforce request/response policies at the gateway layer for security and compliance

GoModel MCP server FAQ

What is GoModel?

GoModel is a self-hosted AI gateway that aggregates MCP servers and LLM providers (OpenAI, Anthropic, Cohere, Google Gemini, etc.) behind one authenticated HTTP endpoint. It provides caching, cost tracking, failover, and observability features.

Is GoModel free?

GoModel is open-source and free to self-host. You pay only for the underlying LLM provider APIs you use. A live demo is available at demo.enterpilot.io.

How do I install GoModel?

Install via curl (macOS/Linux), PowerShell (Windows), or Docker. Run `curl -fsSL https://gomodel.enterpilot.io/install.sh | sh` on Unix, or use `docker run --rm -p 8080:8080 enterpilot/gomodel` for Docker. Access the dashboard at http://localhost:8080/admin/dashboard.

What authentication does GoModel require?

GoModel uses API keys (set via GOMODEL_MASTER_KEY environment variable or generated in the dashboard). It accepts both `Authorization: Bearer` and `x-api-key` headers. Upstream LLM provider credentials are configured via environment variables or the dashboard.

Can I use GoModel with Cursor or Claude?

GoModel exposes OpenAI-compatible and Anthropic-compatible APIs, so you can point any official SDK at your GoModel server by setting the base URL and API key. This works with Cursor, Claude, and any tool using these SDKs.

What LLM providers does GoModel support?

GoModel supports OpenAI, Anthropic, Cohere, Google Gemini, Vertex AI, DeepSeek, Groq, Fireworks AI, Meta, OpenRouter, xAI (Grok), Azure OpenAI, Ollama, vLLM, Amazon Bedrock, and all OpenAI-compatible providers, plus ElevenLabs for voice.

README (reference)

Source of truth, from the repository.

<img alt="GoModel logo" src="docs/logo.svg" width="96"> <h1 align="center"> GoModel - The last AI gateway you will ever need </h1> <a href="https://github.com/ENTERPILOT/GoModel/actions/workflows/test.yml"><img alt="CI" src="https://github.com/ENTERPILOT/GoModel/actions/workflows/test.yml/badge.svg"></a> <a href="https://github.com/ENTERPILOT/GoModel/blob/main/go.mod"><img alt="GO Version" src="https://img.shields.io/github/go-mod/go-version/ENTERPILOT/GoModel?label=GO"></a> <a href="https://hub.docker.com/r/enterpilot/gomodel"><img alt="Docker Pulls" src="https://img.shields.io/docker/pulls/enterpilot/gomodel?label=Docker%20Pulls"></a> <a href="https://discord.gg/gaEB9BQSPH"><img alt="Discord" src="https://img.shields.io/badge/Discord-Join-5865F2?logo=discord&logoColor=white"></a> <a href="https://news.ycombinator.com/item?id=47849097"><img alt="Hacker News" src="https://img.shields.io/badge/Hacker%20News-Apr%2021%20%2726%20%7C%20%234-brightgreen?logo=ycombinator&logoColor=white"></a> <a href="https://gomodel.enterpilot.io/docs?utm_source=readme"><img alt="docs GoModel" src="https://img.shields.io/badge/Docs-GoModel-blue"></a> <a href="https://news.ycombinator.com/item?id=47849097"><img alt="GoModel on Hacker News" src="https://hackerbadge.vercel.app/api?id=47849097"></a> GoModel is the fastest and the most resource-efficient AI Gateway (<a href="https://gomodel.enterpilot.io/docs/about/benchmarks?utm_source=readme">the self-reproducible benchmarks</a>). It's an alternative to LiteLLM (which was hacked recently) and Portkey (which is no longer maintained on GitHub). <a href="https://demo.enterpilot.io/admin/dashboard?utm_source=readme"> <img src="docs/2026-07-07_demo.gif" alt="GoModel AI gateway dashboard showing AI usage analytics, observability panel, token and costs tracking, and estimated cost monitoring" width="100%"> </a> (click on the animation ↑ to see the live demo) GoModel saves you money and nerves. Money - because you can remember the responses on this layer (caching), track your spending and do tricks like prompt compression and intelligent routing. Nerves - because we strive to achieve good quality and reliability. Our ambition is to be the last AI gateway you will need - the most reliable, resource-optimal, feature-rich and fast.

Quick Start

Step 1: Install and start GoModel

macOS / Linux

curl -fsSL https://gomodel.enterpilot.io/install.sh | sh
# OPENAI_API_KEY="your-openai-key" # (optional)
gomodel

Windows (PowerShell)

irm https://gomodel.enterpilot.io/install.ps1 | iex
# $env:OPENAI_API_KEY = "your-openai-key" # (optional)
gomodel

Docker

docker run --rm -p 8080:8080 \
  -e OPENAI_API_KEY="your-openai-key" \
  enterpilot/gomodel

ℹ️ You can configure GoModel with .env variables, a config.yaml file, OR directly in the dashboard.

ℹ️ Full list of environment variables (including all available providers): .env.template

ℹ️ The most secure way in production is to use .env to load API keys.

Step 2: Open the dashboard

http://localhost:8080/admin/dashboard

Step 3: Make your first API call

curl http://localhost:8080/v1/responses \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5-chat-latest",
    "input": "Hello!"
  }'

Using GoModel with official SDKs

GoModel exposes an OpenAI-compatible API at /v1 and an Anthropic-compatible API at /v1/messages, so the official SDKs work unchanged - just point the base URL at your GoModel server and use your GoModel key (set up with the GOMODEL_MASTER_KEY env variable or one generated in the dashboard).

OpenAI SDK

Python

from openai import OpenAI

client = OpenAI(
    base_url="http://localhost:8080/v1",  # your GoModel server
    api_key="your-gomodel-key",
)

TypeScript / JavaScript

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "http://localhost:8080/v1", // your GoModel server
  apiKey: "your-gomodel-key",
});

Anthropic SDK

The Anthropic SDK authenticates with x-api-key, which GoModel accepts alongside Authorization: Bearer.

Python

from anthropic import Anthropic

client = Anthropic(
    base_url="http://localhost:8080",  # your GoModel server (no /v1 suffix)
    api_key="your-gomodel-key",
)

TypeScript / JavaScript

import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic({
  baseURL: "http://localhost:8080", // your GoModel server (no /v1 suffix)
  apiKey: "your-gomodel-key",
});

Supported LLM Providers

GoModel supports OpenAI, Anthropic, Cohere, Google Gemini, Vertex AI, DeepSeek, Groq, Fireworks AI, Meta (Muse Spark), OpenRouter, Z.ai, xAI (Grok), Alibaba Cloud Model Studio (Bailian), Kilo AI, MiniMax, Xiaomi MiMo, OpenCode Go, Azure OpenAI, Oracle, Ollama, SGLang, vLLM, llm-d, Amazon Bedrock Runtime, Amazon Bedrock Mantle, and all OpenAI-compatible providers. Voice: ElevenLabs (text-to-speech and speech-to-text).

See the Providers Overview for the full per-provider feature matrix (chat, /responses, embeddings, files, batches, passthrough), credentials, and configuration notes.

Docker Compose

Infrastructure only (Redis, PostgreSQL, MongoDB, Adminer - no image build):

cp .env.template .env
# Add your API keys to .env
docker compose up -d
# or: make infra

Full stack (adds GoModel + Prometheus; builds the app image):

docker compose --profile app up -d
# or: make image

Service	URL
GoModel API	http://localhost:8080
Adminer (DB UI)	http://localhost:8081
Prometheus	http://localhost:9090

Building the Docker Image Locally

docker build -t gomodel .
docker run --rm -p 8080:8080 --env-file .env gomodel

API Endpoints

GoModel exposes OpenAI-compatible and Anthropic-compatible APIs, provider-native passthrough, and operations routes. See the API Endpoints reference for the full endpoint tables, and Admin Endpoints for the admin REST API and dashboard.

Gateway Configuration

GoModel is configured through environment variables and an optional config.yaml. Environment variables override YAML values. See the Configuration reference for the full list of settings organized by category, along with .env.template and config/config.example.yaml.

Features

Caching - exact and semantic response caching, so repeated prompts cost nothing
Cost tracking - per-request cost estimates, usage analytics, and spending breakdowns in the dashboard
Budgets - hard spend limits per user, team, or key
Rate limits - requests, tokens, and concurrency caps per user path, provider, or model
Virtual models - aliases and load balancing (round-robin or cost-based) behind stable model names
Failover - automatic rerouting to backup providers, with retries and circuit breakers
Labelling - tag requests from HTTP headers or API keys and break down usage by label
User paths - hierarchical scoping of keys, model access, budgets, usage, and audit logs
MCP gateway - aggregate your MCP servers behind one authenticated endpoint
Passthrough API - provider-native APIs under /p/{provider}/..., with GoModel auth and tracking
Guardrails - request and response policies enforced at the gateway
Provider key rotation - round-robin over multiple API keys to lift per-key rate limits
Observability - Prometheus metrics, audit logs, and live request streaming in the dashboard

Roadmap

See the Roadmap for commercial features and the public 0.2.0 milestone.

Community

View repository →