Skip to content
MCP server

LLM Gateway

Usage, cost and model-ranking lookups, chat and image generation through LLM Gateway

Install
Pricing
Open source
Platforms
API, self-hosted, Web
Runs
Remotely
Version
1.0.0
Listing updated
2 days ago

About LLM Gateway

LLM Gateway is an open-source MCP server for ai models and llm tooling. It runs as a hosted remote server, so there is nothing to install locally.

LLM Gateway is an OpenAI-compatible gateway that routes requests to many model providers, and its MCP server turns the account side of that into tools. An assistant can inspect the connected user, organization and API key spending limits, pull request, token, error and cache-hit totals with a daily or hourly time series, and rank providers, models, coding apps or API keys by requests, cost or tokens.

It also sends chat completions to any supported model, generates images with models such as Qwen Image and Gemini 3 Pro Image, and lists available models with capabilities and pricing. The same API key authorises all of it, sent as a bearer token to a stateless Streamable HTTP endpoint; DevPass keys also work, using model ids without a provider prefix.

The gateway itself is open source under AGPLv3 with a separate enterprise licence, and can be self-hosted with Docker or used hosted at llmgateway.io. Usage reports cover up to 366 days, with hourly granularity limited to 31 days.

Install

Run in a terminal

claude mcp add --transport http llmgateway https://api.llmgateway.io/mcp

Add to Settings → Connectors → Add custom connector

https://api.llmgateway.io/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "llmgateway": {
      "url": "https://api.llmgateway.io/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "llmgateway": {
      "type": "http",
      "url": "https://api.llmgateway.io/mcp"
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "llmgateway": {
      "serverUrl": "https://api.llmgateway.io/mcp"
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "llmgateway": {
      "type": "http",
      "url": "https://api.llmgateway.io/mcp"
    }
  }
}

Tools this server provides (3)

Taken from the project documentation; the running server may differ.

get-account
Inspect user, organization, project, role and key spending limits
get-usage
Request and token totals, costs and a time series
get-usage-breakdown
Rank providers, models, apps or API keys by requests, cost or tokens

What it does

Usage analytics
Totals, errors, cache hits, costs and a time series over a date range.
Ranked breakdowns
Rank providers, models, apps or keys by requests, cost or tokens.
Chat and images
Send prompts to any supported model and generate images.
Model discovery
Lists models with capabilities and pricing.
Account view
Shows organization, project, role and key spending limits.
Stateless HTTP
Bearer-key Streamable HTTP endpoint, no session state.

Good for

  • Find which model costs the most this month
  • See which coding app burns the most tokens
  • Check an API key's spending limit
  • Compare model prices before choosing
  • Generate an image from an assistant

Pricing

Open source

Core under AGPLv3 with a separate enterprise licence; hosted pricing not listed

Prices change. Check docs.llmgateway.io before you buy.

LLM Gateway FAQ

How do I authenticate?

Send your LLM Gateway API key as a bearer token; a DevPass plan key also works.

What date range can I query?

Up to 366 days, and hourly reports allow at most 31 days.

Can I self-host the gateway?

Yes. The README documents a Docker setup; the core is AGPLv3.

Which endpoint do I use?

https://api.llmgateway.io/mcp, or https://llmgateway.io/mcp.

Is LLM Gateway free?

Yes. LLM Gateway is open source, so you can use and self-host it for free.

How do I add LLM Gateway to Claude Code?

Run this command in your terminal: claude mcp add --transport http llmgateway https://api.llmgateway.io/mcp

What are the best alternatives to LLM Gateway?

The closest alternatives to LLM Gateway are Claude Code Ultimate Guide, signafAIde, Toolkien, Lugh and HypePaper.

Alternatives to LLM Gateway

  1. MCP server for searching the Claude Code Ultimate Guide from any MCP client

    Runs locally
    • Open source
    • 6.5K GitHub stars
    • AI models and LLM tooling
  2. signafAIde

    MCP server

    Local MCP server to set up, verify and check coverage of signed agent action records

    Runs locally
    • 5 tools
    • Freemium
    • Security
  3. Toolkien

    MCP server

    One MCP gateway for all your MCP servers, with code mode, permissions and audit

    Not checked yet
    • 3 tools
    • Remote
    • Open source
    • AI models and LLM tooling
  4. Lugh

    MCP server

    Write, check and dry-run Lugh agent programs with policy, budgets and approvals

    Runs locally
    • 5 tools
    • Open source
    • AI models and LLM tooling
  5. HypePaper

    MCP server

    Research papers, benchmarks, citations and topic experts for AI assistants

    • 58 tools
    • Remote
    • Freemium
    • AI models and LLM tooling