Evaluation, Observability & Safety

OpenLIT

openlit/openlit

OpenTelemetry-native observability and evaluation tools for LLM applications and AI infrastructure.

★ 2.8KStars
⑂ 423Forks
74Open issues
PythonLanguage
Apache-2.0License
Q84Editorial score

Overview

OpenLIT instruments LLM applications and supporting infrastructure with OpenTelemetry-compatible traces and metrics. Its features cover prompt and model activity, evaluation, guardrails, cost visibility, and GPU monitoring.

Key features

  • OpenTelemetry-compatible tracing
  • LLM and agent instrumentation
  • Evaluation and guardrail features
  • Cost and GPU visibility

Requirements, installation and quick start

Follow the official quickstart to install the SDK or deploy the observability platform; choose the integration for your framework.

Usage

Instrument an application, send telemetry to the configured collector or platform, then inspect traces and evaluation signals.

Model compatibility and use cases

Integrations cover multiple LLM frameworks and providers; check the current documentation for supported versions.

License and risk notes

The repository declares Apache-2.0. Telemetry backends and connected providers may have separate terms.

Release and maintenance

See the official repository for current releases and changes.

netdata

netdata/netdata

★ 80.4KGo

strix

usestrix/strix

★ 60.6KPython

open_spiel

google-deepmind/open_spiel

★ 5.4KC++