Observability, Evaluations, Rule Engine, Guardrails, Prompts, Vault, Playground

Open Source Platform for AI Engineering

Documentation | Quickstart | Python SDK | Typescript SDK | Go SDK |

❤️ Sponsor this project ❤️

https://github.com/user-attachments/assets/6909bf4a-f5b4-4060-bde3-95e91fa36168

OpenLIT allows you to simplify your AI development workflow, especially for Generative AI and LLMs. It streamlines essential tasks like experimenting with LLMs, organizing and versioning prompts, and securely handling API keys. With just one line of code, you can enable OpenTelemetry-native observability, offering full-stack monitoring that includes LLMs, vector databases, and GPUs. This enables developers to confidently build AI features and applications, transitioning smoothly from testing to production.

This project proudly follows and maintains the Semantic Conventions with the OpenTelemetry community, consistently updating to align with the latest standards in Observability.

⚡ Features

• 📈 Analytics Dashboard: Monitor your AI application's and GPUs health and performance with detailed dashboards that track metrics, costs, and user interactions, providing a clear view of overall efficiency.

• 🔌 OpenTelemetry-native Observability SDKs: Vendor-neutral SDKs (Python, TypeScript, Go) to send traces and metrics to your existing observability tools.

• 🛡️ 11 Built-in Evaluation Types: Automated LLM-as-a-Judge evaluation with hallucination, bias, toxicity, safety, instruction following, completeness, conciseness, sensitivity, relevance, coherence, and faithfulness detection. Context-aware evaluation that treats provided context as the source of truth.

• ⚙️ Rule Engine: Define conditional rules with AND/OR logic to match runtime trace attributes and dynamically retrieve contexts, prompts, and evaluation configs. SDK support across Python, TypeScript, and Go.

• 💲 Cost Tracking for Custom and Fine-Tuned Models: Tailor cost estimations for specific models using custom pricing files for precise budgeting.

• 🐛 Exceptions Monitoring Dashboard: Quickly spot and resolve issues by tracking common exceptions and errors with a dedicated monitoring dashboard.

• 💭 Prompt Management: Manage and version prompts using Prompt Hub for consistent and easy access across applications.

• 🔑 API Keys and Secrets Management: Securely handle your API keys and secrets centrally, avoiding insecure practices.

• 🎮 Experiment with different LLMs: Use OpenGround to explore, test and compare various LLMs side by side.

🚀 Getting Started with LLM Observability

flowchart TB;
    subgraph " "
        direction LR;
        subgraph " "
            direction LR;
            OpenLIT_SDK[OpenLIT SDK] -->|Sends Traces & Metrics| OTC[OpenTelemetry Collector];
            OTC -->|Stores Data| ClickHouseDB[ClickHouse];
        end
        subgraph " "
            direction RL;
            OpenLIT_UI[OpenLIT] -->|Pulls Data| ClickHouseDB;
        end
    end

Step 1: Deploy OpenLIT Stack

1. Git Clone OpenLIT Repository

   Open your command line or terminal and run:

   git clone git@github.com:openlit/openlit.git
   

2. Self-host using Docker

   Deploy and run OpenLIT with the following command:

   docker compose up -d
   

For instructions on installing in Kubernetes using Helm, refer to the Kubernetes Helm installation guide.

Step 2: Install OpenLIT SDK

Open your command line or terminal and run:

pip install openlit

For instructions on using the TypeScript SDK, visit the TypeScript SDK Installation guide.

Step 3: Initialize OpenLIT in your Application

Integrate OpenLIT into your AI applications by adding the following lines to your code.

import openlit

openlit.init()

Configure the telemetry data destination as follows:

💡 Info: If the otlp_endpoint or OTEL_EXPORTER_OTLP_ENDPOINT is not provided, the OpenLIT SDK will output traces directly to your console, which is recommended during the development phase.

Example

Add the following two lines to your application code:

import openlit

openlit.init(
  otlp_endpoint="http://127.0.0.1:4318", 
)

Add the following two lines to your application code:

import openlit

openlit.init()

Then, configure the your OTLP endpoint using environment variable:

export OTEL_EXPORTER_OTLP_ENDPOINT = "http://127.0.0.1:4318"

Step 4: Visualize and Optimize

With the Observability data now being collected and sent to OpenLIT, the next step is to visualize and analyze this data to get insights into your AI application's performance, behavior, and identify areas of improvement.

Just head over to OpenLIT at 127.0.0.1:3000 on your browser to start exploring. You can login using the default credentials:

• Email: user@openlit.io
• Password: openlituser

🤖 AI Coding Agents (Claude Code, Cursor, Codex)

OpenLIT also ships a CLI that brings the same observability story to your local coding agents — Claude Code, Cursor, and Codex. No SDK to import: the CLI installs vendor hooks that emit OpenTelemetry traces for every session, prompt, tool call, file edit, subagent spawn, and code-impact event.

Install the CLI

# macOS + Linux
curl -fsSL https://raw.githubusercontent.com/openlit/openlit/main/cli/scripts/install.sh | sh

# Windows (PowerShell)
iwr -useb https://raw.githubusercontent.com/openlit/openlit/main/cli/scripts/install.ps1 | iex

Wire it into your agents

openlit configure --endpoint http://127.0.0.1:4318 [--api-key <key>]
openlit coding install --vendor=all   # or: cursor / claude-code / codex

Inspect what's running

openlit doctor    # config + OTLP reachability + installed plugins

Sessions, prompts, costs, and per-vendor breakdowns appear in the OpenLIT dashboard at http://127.0.0.1:3000/coding-agents. To detach later:

openlit coding uninstall --vendor=all

Coding-agent observability is OTel-native end-to-end: traces use the gen_ai.* semantic conventions alongside an coding_agent.* extension namespace. Anything that speaks OTLP (Datadog, Honeycomb, Grafana Tempo, raw OTel Collector) can consume the same data — OpenLIT is just one possible viewer.

📦 Supported Integrations

OpenLIT auto-instruments 50+ LLM providers, AI frameworks, and vector databases with a single line of code. Each integration produces OpenTelemetry-native traces and metrics. Click any card to view the integration docs.

LLM Providers