Spanlens is an open-source (MIT) LLM observability platform that logs every OpenAI, Anthropic, and Gemini request in one line of code.
- Track cost, latency, and token usage per request
- Trace multi-step agent workflows as waterfall span trees
- Detect anomalies and PII (SSN, email, etc.) with automatic API‑key masking
- Recommend cheaper models with dollar‑figure savings estimates
- Provide LLM‑as‑judge evaluations (0‑1 scores) to measure prompt quality
- Support self‑hosted deployment via Docker Compose or a single binary
Ideal for developers and teams building LLM‑powered applications who need visibility into cost, performance, and safety.
