Local-first LLM API cost tracker that monitors usage across Anthropic, OpenAI, Qwen, and DeepSeek providers, storing data in SQLite and providing spend analysis, provider comparisons, and optimization recommendations.
zhaoyue722/llm-usage-mcp is an MCP (Model Context Protocol) server that tracks and analyzes LLM API costs in real-time. It captures usage data from multiple providers including Anthropic, OpenAI, Qwen, and DeepSeek, maintaining a local SQLite database ledger. The tool provides spend queries, detailed provider comparisons, and actionable cost optimization recommendations with first-class support for Chinese providers and CNY to USD currency conversion.
Installation requires Python 3.7+ and pip. Clone the repository from GitHub, install dependencies via pip, configure your API keys for supported providers in environment variables, and initialize the local SQLite database. The MCP server can then be registered with compatible clients like Claude Desktop or other MCP-compatible applications to expose cost tracking as integrated tools.
Monday.com MCP Server streamlines board management, item operations, and workflow automation for teams. I…
от NotionFlow
Sentry MCP Server provides comprehensive error tracking and performance monitoring, helping developers id…
от AnalyticsPro
Cloudflare MCP Server simplifies Cloudflare management by providing tools for DNS management, Workers dep…
от PricingBot