nikhilnt1234/TokenBurnRate

Overview

TokenBurnRate tracks LLM token costs across models from Claude, GPT, and Gemini. It delivers an MCP server and a CLI, with optimization hints and estimated cost savings.

Best for

Best for
Developers who need to monitor and optimize LLM API costs across multiple providers

Use cases

• Monitor token spend across multiple LLM providers in real time
• Identify cost-saving opportunities from usage patterns
• Integrate cost tracking into development workflows via MCP or CLI

How to use

Install

npx token-tracker-mcp report

Tools exposed

• log_usage
• get_summary
• get_hints
• get_hint_detail
• set_budget
• list_sessions
• list_models
• export_csv
• cache-utilization
• model-swap-testing
• model-swap-debug
• verbose-outputs
• session-spike
• context-bloat
• retry-loops
• single-model-dependency

Tested with

Claude Desktop

Notes

TokenBurnRate tracks LLM token costs across models from Claude, GPT, and Gemini. It delivers an MCP server and a CLI, with optimization hints and estimated cost savings.

0 stars on GitHub. Last updated 2026-06-03. Licensed MIT.

Use cases

• Monitor token spend across multiple LLM providers in real time
• Identify cost-saving opportunities from usage patterns
• Integrate cost tracking into development workflows via MCP or CLI

Pros

• Supports three major LLM providers (Claude, GPT, Gemini)
• Provides both an MCP server and a CLI for flexible integration
• Offers actionable optimization hints and cost savings estimates

Cons

• Low community adoption (0 GitHub stars) indicates limited validation
• Focused solely on cost tracking, no usage management or budgeting features
• Requires manual setup and configuration for each provider

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.