Spanlens

Drop-in LLM observability logging OpenAI, Anthropic, and Gemini requests

PlatformWebfreemiumglobal

Spanlens is an MIT-licensed LLM observability platform giving full visibility into every request an app makes to OpenAI, Anthropic, or Gemini. It's a drop-in replacement for the provider SDK — swap one import and see data immediately, no separate infra to run. It logs full request bodies/headers/tokens/cost/latency, tracks per-model and per-route cost with budget alerts via Slack/webhook, renders multi-step agent workflows as waterfall span trees with cost attribution, flags anomalies against a rolling baseline, auto-masks API keys in logged request bodies, projects savings from switching to cheaper models, and supports LLM-as-judge scoring to compare prompt versions. Supports OpenAI, Anthropic, Gemini, Mistral, Azure, Bedrock, Vertex, and OpenTelemetry ingestion; fully self-hostable with export to JSON/CSV/Parquet or S3/BigQuery.

Website

Full match profile

Behind the summary, Matchbox keeps a richer profile of Spanlens - the signals our matcher actually reads to decide when to surface it. It stays private; claim the listing to see and control it.

Problem & pain-point mapping
Who we surface it to (audience fit)
What it's a strong alternative to
Trust & credibility signals

Claim Spanlens to unlock →

Try Matchbox with your own problem

Describe what is not working - we’ll show you whether Spanlens (or something else) actually fits.

Try a problem