📄 中文摘要
llm-sentry 是刚刚在 PyPI 上发布的 Python 包，专注于 LLM 管道监控、故障诊断和合规检查。对于在生产环境中运行 AI 的开发者来说，这是必不可少的工具。然而，监控只是解决方案的一部分，还需要一个可靠且具有成本效益的推理后端来实际调用模型，这正是 NexaAPI 的作用。该教程展示了如何将 llm-sentry 的监控能力与 NexaAPI 提供的 56 种以上模型推理 API 结合，构建一个完整的生产 LLM 栈。
llm-sentry + NexaAPI：完整的 LLM 可靠性栈仅需 10 行代码
📄 English Summary
llm-sentry + NexaAPI: The Complete LLM Reliability Stack in 10 Lines of Code
llm-sentry has just been released on PyPI as a Python package for monitoring LLM pipelines, diagnosing faults, and ensuring compliance. This tool is essential for developers running AI in production environments. However, monitoring is only part of the solution; a reliable and cost-effective inference backend is also needed to call the models, which is where NexaAPI comes into play. This tutorial demonstrates how to combine the monitoring capabilities of llm-sentry with NexaAPI's 56+ model inference API to create a complete production LLM stack.