Introducing W&B Weave

A visual, interactive, and composable way of building ML & data apps. Build and share in minutes. Python and no-code unified.

A new way to build machine learning and data apps

W&B Weave is a visual and interactive toolkit for machine learning and data app development. Start with a table of data, end with any interactive dashboard you can dream up, easily switching between code and UI along the way. Transform the way you understand, visualize and interact with your data.

Composable and extensible with a rich ecosystem

Weave lets you build and extend modular components, so you can start simply and add sophistication along the way. You can share and reuse every step, leveraging your own work and the ML ecosystem’s newest techniques to do everything from exploring your data to comparing model configurations.

Intuitive and performant ML workflows

Weave is designed specifically to compose machine learning and data applications. Intuitively compose end-to-end ML workflows, including data exploration, experiments, evaluation, monitoring, and more.

Versatile and adaptive to your domain

We built Weave for both tool creators and consumers. It’s customizable for your unique use case and problem domain, allowing you to build and run performant, reliable apps in production for any audience or stakeholder.
Works with 19,000+ ML libraries & repos

Interoperable with anything

Weave integrates easily, with support for the data analysis tools and repos you’re using already. And it takes just two lines of code to get started.

Get started in seconds

You can get started with Weave open source in your notebook with just two lines of code.

import weave

Explore the Weights & Biases platform


Experiment tracking


Collaborative dashboards


Dataset and
model versioning


Interactive data visualization


Hyperparameter optimization


Automate ML workflows


Model lifecycle management


Observability for production ML


LLMOps and prompt engineering


ML app builder