langtrace
Langtrace is an open source observability tool for LLM applications. It captures, debugs, and analyzes traces and metrics from applications using LLM APIs, vector databases, and LLM frameworks. It is designed for developers who want to monitor performance, costs, and usage of their LLM-based systems.
✨ Key features
- Built on OpenTelemetry standards for comprehensive tracing.
- Real-time monitoring of LLM API calls, vector operations, and frameworks.
- Analyze latency, costs, and usage patterns.
- Trace and debug LLM application workflows.
- Detailed metrics and visualizations.
- Self-hosting option for deployment on own infrastructure.
🎯 Use cases
- Monitor and debug LLM application workflows.
- Track performance and cost of LLM API calls.
- Analyze usage patterns across different LLM providers and frameworks.
- Self-host observability for privacy-sensitive LLM applications.
📦 Installation
🧰 Requirements: Requires Docker and Docker Compose for self-hosting; SDKs require TypeScript/JavaScript or Python runtime. For cloud, an API key from langtrace.ai is needed.
For TypeScript/JavaScript
npm i @langtrase/typescript-sdk
For Python
pip install langtrace-python-sdk
Self-hosting (Docker)
docker compose up
The application will be available at http://localhost:3000.
🚀 Usage
TypeScript
import * as Langtrace from '@langtrase/typescript-sdk' // Must precede any llm module imports
Langtrace.init({ api_key: '<your_api_key>' }) // Get your API key at langtrace.ai
Python
from langtrace_python_sdk import langtrace
langtrace.init(api_key='<your_api_key>') # Get your API key at langtrace.ai
⚠️ Good to know
The semantic conventions for traces are under ongoing development. The project is open source and welcomes feedback.
❓ FAQ
Can I self-host Langtrace?
Yes, you can self-host Langtrace by following the self-hosting setup instructions in the documentation. It requires Docker and Docker Compose.
What is the pricing for Langtrace cloud?
Currently, Langtrace cloud is free. The team is primarily looking for feedback and will inform users when they decide to monetize.
What is the tech stack of Langtrace?
Langtrace uses NextJS for the frontend and APIs, PostgresDB as a metadata store, and Clickhouse DB for storing spans, metrics, logs, and traces.
Can I contribute to this project?
Absolutely! Contributions are welcome. You can get involved by joining the Discord community and forking the repository.
📊 Repository
🤖 Overview, features, install steps and FAQ were generated from the project's README on Sep 4, 2026. Always check the original source before running commands.