# GoModel

> GoModel is an MCP server for ai models and llm tooling. Self-hosted AI gateway with caching, budgets, failover and an MCP aggregation endpoint. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Platforms:** Linux, macOS, Windows, CLI, self-hosted, API
- **Repository:** https://github.com/enterpilot/gomodel
- **MCP Registry name:** io.github.ENTERPILOT/gomodel
- **Categories:** [AI models and LLM tooling](https://ai-powered.com/categories/ai-and-llm), [developer tools](https://ai-powered.com/categories/developer-tools)
- **Listing updated:** 2026-11-15

## About

The headline product is an LLM gateway that accepts OpenAI-compatible requests at /v1 and Anthropic-compatible requests at /v1/messages, so official SDKs work after changing a base URL. It routes to a long provider list including OpenAI, Anthropic, Gemini, Vertex AI, Bedrock, Azure OpenAI, Mistral, Groq, Ollama and vLLM.

Gateway features include exact and semantic caching, cost tracking, budgets and rate limits by user path, virtual models with load balancing, session pinning, failover with circuit breakers, guardrails and plugins, with a dashboard at /admin/dashboard. Its MCP piece aggregates upstream MCP servers behind one authenticated endpoint, configured with a MASTER_KEY and an MCP_SERVERS JSON object.

It runs from a Docker image or an install script. The registry notes that leaving the master key unset runs the gateway with no authentication.

## Features

- **OpenAI and Anthropic APIs:** Accepts both request formats so SDKs need only a new base URL.
- **Many providers:** Routes to dozens of hosted and local model providers.
- **Caching and budgets:** Exact and semantic caching with spend limits and cost tracking.
- **Failover:** Reroutes to backup providers with retries and circuit breakers.
- **MCP gateway:** Aggregates upstream MCP servers behind one authenticated endpoint.
- **Admin dashboard:** Manages providers and settings in a web UI.

## Good for

- Put one gateway in front of several LLM providers
- Cap spend per team or key
- Expose several MCP servers through one authenticated URL
- Keep provider prompt caches warm with session pinning

## Install

### Claude Code (Terminal)

```sh
claude mcp add gomodel --env GOMODEL_MASTER_KEY='<GOMODEL_MASTER_KEY>' -- docker run -i --rm -e GOMODEL_MASTER_KEY docker.io/enterpilot/gomodel:0.2.5
```

### Claude Desktop (claude_desktop_config.json)

```json
{
  "mcpServers": {
    "gomodel": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "GOMODEL_MASTER_KEY",
        "docker.io/enterpilot/gomodel:0.2.5"
      ],
      "env": {
        "GOMODEL_MASTER_KEY": "<GOMODEL_MASTER_KEY>"
      }
    }
  }
}
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "gomodel": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "GOMODEL_MASTER_KEY",
        "docker.io/enterpilot/gomodel:0.2.5"
      ],
      "env": {
        "GOMODEL_MASTER_KEY": "<GOMODEL_MASTER_KEY>"
      }
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "gomodel": {
      "type": "stdio",
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "GOMODEL_MASTER_KEY",
        "docker.io/enterpilot/gomodel:0.2.5"
      ],
      "env": {
        "GOMODEL_MASTER_KEY": "<GOMODEL_MASTER_KEY>"
      }
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "gomodel": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "GOMODEL_MASTER_KEY",
        "docker.io/enterpilot/gomodel:0.2.5"
      ],
      "env": {
        "GOMODEL_MASTER_KEY": "<GOMODEL_MASTER_KEY>"
      }
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "gomodel": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "GOMODEL_MASTER_KEY",
        "docker.io/enterpilot/gomodel:0.2.5"
      ],
      "env": {
        "GOMODEL_MASTER_KEY": "<GOMODEL_MASTER_KEY>"
      }
    }
  }
}
```

## FAQ

### How do I run it?

With Docker (docker run enterpilot/gomodel) or the install script.

### What formats does it accept?

OpenAI-compatible at /v1 and Anthropic-compatible at /v1/messages.

### What if no master key is set?

The gateway runs in unsafe no-auth mode.

### Where is the dashboard?

At /admin/dashboard on the server.

### How do I add GoModel to Claude Code?

Run this command in your terminal: claude mcp add gomodel --env GOMODEL_MASTER_KEY='<GOMODEL_MASTER_KEY>' -- docker run -i --rm -e GOMODEL_MASTER_KEY docker.io/enterpilot/gomodel:0.2.5

### What are the best alternatives to GoModel?

The closest alternatives to GoModel are ByMorning, LLM Gateway, codai, MCP-Bridge and Unified AI System MCP Gateway.

## Where this comes from

- Description: written from the project's README.
- Install steps: from the official MCP Registry entry.

## Alternatives

- [ByMorning](https://ai-powered.com/mcp/bymorning): ByMorning is an open-source MCP server for ai models and llm tooling.
- [LLM Gateway](https://ai-powered.com/mcp/llm-gateway): LLM Gateway is an open-source MCP server for ai models and llm tooling.
- [codai](https://ai-powered.com/mcp/codai): codai is an MCP server for ai models and llm tooling.
- [MCP-Bridge](https://ai-powered.com/mcp/mcp-bridge-2): MCP-Bridge is an open-source MCP server for ai models and llm tooling.
- [Unified AI System MCP Gateway](https://ai-powered.com/mcp/unified-ai-system-mcp-gateway): Unified AI System MCP Gateway is an open-source MCP server for developer tools.
- [Opper](https://ai-powered.com/mcp/opper): Opper is a freemium MCP server for ai models and llm tooling.
- [RelaySwitch](https://ai-powered.com/mcp/relayswitch-2): RelaySwitch is a paid MCP server for ai models and llm tooling.
- [Local AI MCP](https://ai-powered.com/mcp/local-ai-mcp): Local AI MCP is an open-source MCP server for ai models and llm tooling.

---
Source: https://ai-powered.com/mcp/gomodel
