TokenBurnRate MCP server

TokenBurnRate is a small helper that keeps a record of every AI call you make and turns it into a simple cost report. It shows you where your money goes and gives you clear tips on how to lower the bill. It is handy if you use Claude, GPT or Gemini often and want to know what it actually costs.

What this MCP server does

You ask your AI something like how much you spent this week. The AI passes that question to TokenBurnRate, which reads its local log of your API calls. It adds up the tokens and the cost, checks for patterns like unused caching or expensive models doing simple work, and hands the answer back. You see the total, the daily breakdown and a short list of saving tips right in the chat.

Click to zoom

What you can do with it

• See your total AI cost for today, this week or this month
• Get a ranked list of saving tips with estimated monthly savings
• Set a daily, weekly or monthly spending limit
• List your sessions ranked by how much they cost
• Look up the price table for all supported models
• Export your raw usage data as a CSV file
• Log a call manually and let it work out the cost

Try asking your AI

• “How much did I spend on AI this week?”
• “Show me my top three ways to save money on tokens”
• “Set a monthly budget of 50 dollars and warn me when I get close”
• “Which of my sessions cost the most this month?”

What it gives back to you

You get numbers and short lists in the chat: total cost, number of calls, token counts and a daily breakdown. The saving tips come as short items with a severity, the evidence and an estimated monthly saving. If you ask for a CSV, it gives you a file you can open in a spreadsheet.

Before you start

Who it's for

Anyone who uses Claude, GPT or Gemini regularly and wants to keep an eye on what it costs, especially freelancers, small teams and people on a budget.