ModelRefs / Weights & Biases (W&B) — AI Glossary
Weights & Biases (W&B) — AI Glossary
A cloud ML experiment tracking and visualization platform with run comparison, hyperparameter sweeps, and artifact versioning. Also called W&B or wandb.
Overview
W&B provides: Runs (experiment logging with rich visualizations), Sweeps (hyperparameter optimization), Artifacts (dataset/model versioning with lineage), Tables (multi-media data exploration), and Weave (LLM trace tracking and evaluation). Used by OpenAI, DeepMind, and the majority of ML teams for LLM pretraining and fine-tuning.
Reference details
| Topic | infrastructure |
|---|---|
| Also known as | W&B, wandb |
| Last reviewed | 2026-06-24 |
Related terms
Continue your research
Use these connected ModelRefs sections to compare alternatives, inspect implementation paths, and review the evidence and governance boundaries relevant to Weights & Biases (W&B) — AI Glossary.
Frequently asked questions
What is Weights & Biases (W&B)?
A cloud ML experiment tracking and visualization platform with run comparison, hyperparameter sweeps, and artifact versioning.
Is Weights & Biases (W&B) the same as W&B?
Yes — W&B, wandb are common aliases for Weights & Biases (W&B).
What concepts are related to Weights & Biases (W&B)?
Closely related concepts include mlflow, experiment tracking, langsmith.