Spanlens

Spanlens is an open-source (MIT) LLM observability platform that logs every OpenAI, Anthropic, and Gemini request with a single line of code.

• Track cost, latency, and token usage per request
• Trace multi-step agent workflows as waterfall spans
• Detect anomalies (3σ) and PII/secret leakage
• Recommend cheaper models with dollar‑figure savings
• LLM‑as‑judge evals (0‑1) to measure quality improvements
• Experiments & datasets to compare prompt/model versions
• Self‑hostable (Docker, single binary) with data staying in your VPC
• Supports TypeScript, Python, Next.js, Node, Edge, LangChain, LlamaIndex, Vercel AI SDK

Ideal for developers and teams building LLM‑powered apps who need cost visibility, latency debugging, security scanning, and model optimization—all in one place.