> ## Documentation Index
> Fetch the complete documentation index at: https://docs.openlit.io/llms.txt
> Use this file to discover all available pages before exploring further.
# OpenLIT
> Open-source AI engineering platform
OpenLIT is an open-source AI engineering platform that helps teams build, evaluate and observe AI applications across the entire lifecycle from development to production.
OpenLIT provides the following OSS tools to support this goal:
* [OpenLIT](/latest/openlit/installation) - Open-source platform for tracing, prompt management, evaluations, and scalable AI observability with dashboards, metrics, logs, and remote collectors.
* [OpenLIT SDKs](/latest/sdk/overview) - OpenTelemetry-native auto-instrumentation to trace LLMs, agents, vector databases and GPUs with zero-code.
* [OpenLIT Controller](/latest/controller/overview) - Zero-code LLM and Agent observability for Kubernetes, Docker, and Linux using eBPF and automatic SDK injection.
The project proudly follows OpenTelemetry Semantic Conventions for AI observability standards.
## Features
OpenLIT provides distributed tracing capabilities for understanding and debugging AI applications:
* [OpenTelemetry-native SDKs](/latest/sdk/overview) - Automatic instrumentation for LLMs, agents, frameworks, Vector databases, MCP and GPUs.
* [Exceptions Monitoring](/latest/openlit/observability/error) - Track and debug application errors with detailed stack traces.
* **Universal Compatibility** - View traces from any OpenTelemetry-instrumented tool or LLM instrumentation frameworks like OpenInference, OpenLLMetry.
OpenLIT provides automated LLM evaluation and testing capabilities for production AI applications:
* [Online LLM-as-a-jduge](/latest/openlit/evaluations/llm-as-a-judge) - Zero-setup AI quality monitoring that automatically evaluates your LLM responses from traces.
* [Programmatic Evaluations](/latest/openlit/evaluations/programmatic-evals) - Programmatic AI evaluation tools for custom testing workflows and development pipelines.
OpenLIT's [Prompt Hub](/latest/openlit/prompts-experiments/prompt-hub) provides capabilities to centrally manage and maintain prompts:
* Version and edit prompts collaboratively
* Deploy prompts to any environment without code changes
* Track prompt evolution and changes over time
OpenLIT provides experimentation tools for comparing and optimizing AI models:
* [OpenGround](/latest/openlit/prompts-experiments/openground) - Compare cost, duration, and response tokens across LLMs to choose the most efficient model for your use case.
[Dashboards](/latest/openlit/dashboards/overview) in OpenLIT provide powerful visualization capabilities for AI monitoring:
* Create and resize custom widgets with flexible configurations and layouts
* Visualize telemetry from any OpenTelemetry-instrumented tool or LLM instrumentation frameworks like OpenInference, OpenLLMetry
* Write custom SQL queries to analyze your AI telemetry data
* Export and import dashboard configurations as JSON files
OpenLIT provides secure credential management for AI applications:
* [Vault](/latest/openlit/developer-resources/vault) - Centrally store LLM API keys that applications can retrieve remotely without restarts for seamless key rotation.
## Upgrade Information
**Important Upgrade Notice**: When upgrading from OpenLIT versions prior to 1.15.0 (which introduced Fleet Hub), special attention is required for Docker Compose deployments.
If you're running OpenLIT using Docker Compose, you **must** use the `--remove-orphans` flag when upgrading:
```bash theme={null}
# Upgrade with orphan removal
docker-compose up -d --remove-orphans
```
This is necessary because the OpenTelemetry Collector has been integrated directly into the OpenLIT container, and the standalone `otel-collector` container is no longer needed. Without `--remove-orphans`, the old collector container will conflict with the new integrated collector on ports 4317 and 4318.
## Getting Started
Choose your path to start building better AI applications with OpenLIT:
Production-ready AI monitoring setup in 2 simple steps with zero code changes
Deployment options for scalable LLM monitoring infrastructure
Create custom visualizations with flexible widgets, queries, and real-time AI monitoring
Version, deploy, and collaborate on prompts with centralized management and tracking
Discover and instrument LLM traffic across Kubernetes, Docker, and Linux using eBPF — no code changes required.