W&B Weave - Weights & Biases Documentation

W&B Weave is an observability and evaluation platform that helps you track, evaluate, and improve your LLM application. With Weave, you can:

Observe and debug your LLM application
Evaluate your application’s responses using LLM judges and custom scorers

Get Started

The following docs guide you through the basics of how to use Weave’s suite of tools.

Quickstart: Track LLM inputs & outputs

Start by tracing a basic call to an LLM and reviewing the data in your W&B account.

Get started evaluating your app

Learn how to build an evaluation pipeline using Weave scorers to test and track your application’s performance.

Evaluate a RAG application

Build and evaluate RAG applications using Weave with LLM judges to measure retrieval quality.

Install Weave

W&B Weave provides Python and TypeScript libraries. To install the Weave library, run the following command:

Python
TypeScript

pip install weave

To start using the Weave library, create a Weights & Biases (W&B) account and obtain a copy your API key. The API key allows you to authenticate to your W&B account and start sending data to it.

Quickstart: Track LLM inputs & outputs