MCP server · Analytics

LLM Usage & Cost Tracker MCP server

Ask your AI how much you spent on Claude, GPT, Qwen or DeepSeek, and which model is cheapest.

Analytics For it and development Monitoring

This is a small helper that keeps a record of what your AI calls cost. It watches the calls you make to Anthropic, OpenAI, Qwen and DeepSeek, saves the numbers in a file on your own computer, and lets you ask your AI questions about your spending in plain English. It is handy if you use more than one AI provider and want to stop guessing what the monthly bill will be.

What is an MCP server? The 30-second version

On its own, your AI can only chat. It cannot see your bills or your spending history. An MCP server is a small helper program that gives your AI a new skill, and this one gives it a connection to your own usage records. Once it is connected, your AI can look up what you spent, compare providers, and suggest cheaper models when you ask.

What this MCP server does

You ask your AI something like how much you spent on Claude this week. Your AI uses this helper to look inside a small database file on your computer where every call has been recorded. The helper adds up the tokens and the cost for the time window you asked about, using the current prices for each provider. Then your AI reads the answer back to you in the chat, in plain words and numbers.

Flow diagram: you ask your AI “How much did I spend on Claude this week?”, on your own computer the LLM Usage & Cost Tracker MCP server works with your own computer, and you get back your spending answer.

Click to zoom

What you can do with it

See your total spend for today, this week, this month or this year
Break your spending down by provider, model, project or day
Compare what the same call would cost across every priced model
Find the cheapest model that fits a budget you set
Look up the current price per million tokens for any model
List which providers are set up on your machine
Log a call by hand when it was not captured automatically

Try asking your AI

“How much did I spend on Anthropic today?”
“Show me my spend this month grouped by model.”
“Which provider is cheapest for a 10k input and 2k output call?”
“I have 4 cents left, which model should I use for a 1k by 1k call?”

What it gives back to you

You get back plain answers in the chat: totals in dollars, lists of your top providers and models, and the largest single call. When you ask for a comparison, you get a ranked list of models with their projected cost and how much more expensive each one is than the cheapest. When you ask for a recommendation, you get one model plus two backups and a short note explaining why.

Before you start

What you need

Python 3.10 or newer
The uv tool (or pipx) to install the package
At least one API key for the provider you use, like ANTHROPIC_API_KEY or OPENAI_API_KEY
The capture proxy running so your calls get recorded

Good to know

It only reads and records numbers, it does not send or change your AI calls, but the database file sits on your computer, so keep it private if your usage details are sensitive.

Install it with your AI

Add LLM Usage & Cost Tracker MCP server to your AI, no technical skills needed

You don't install anything by hand. You copy one prompt, paste it into an AI that can work on your computer, and it checks, installs and connects the server for you, asking you when it needs something.

Sign in to get the install prompt

Members get a ready-made prompt that lets the Claude desktop app check LLM Usage & Cost Tracker MCP server, install it and connect it for them, step by step. You don't need any technical skills: you copy, paste and answer a few questions. Your connected AI can also find and install any of the 4,066 MCP servers here for you.