OpenLIT

OpenLIT

OpenTelemetry-native LLM observability — tracing, evaluation, prompt management, and cost tracking

Open source alternative to:Datadog

OpenLIT is an OpenTelemetry-native LLM observability platform with 2.5k+ GitHub stars — a Datadog alternative for tracing, evaluation, prompt management, and cost tracking.

Key features

Observability

  • OpenTelemetry-native architecture
  • Distributed tracing for LLM chains
  • Real-time monitoring and alerting
  • Custom dashboards and metrics
  • Integration with existing OTel collectors

Evaluation

  • Automated quality scoring
  • Regression testing for prompts
  • Side-by-side comparison
  • Human-in-the-loop annotation
  • Custom evaluation metrics

Prompt management

  • Version control for prompts
  • A/B testing and experiments
  • Template management
  • Playground for debugging
  • Rollback and staging

Cost tracking

  • Per-request cost attribution
  • Model and provider comparison
  • Budget alerts and quotas
  • Usage analytics by team/user
  • Export to billing systems

At a glance

LicenseApache-2.0
StackPython, TypeScript, PostgreSQL
Self-hostedYes — OpenLIT OSS
Cloudopenlit.io (managed)
ProtocolOpenTelemetry

Self-hosting

git clone https://github.com/openlit/openlit.git
cd openlit && docker compose up

OpenLIT can be self-hosted with Docker. The cloud version provides managed infrastructure and automatic updates.

Screenshots

OpenLIT screenshot 1

Category

Analytics

Tags

aillmobservabilitymonitoring