Weights & Biases
Free tierML experiment tracking, model evaluation, and LLMOps platform used by leading AI teams
Free tier available·Technical·API available
Key strengths
Industry standard for ML experiment trackingWeave module for LLM tracing and evaluationSweeps for hyperparameter optimizationModel registry with lineage trackingAcquired by CoreWeave — deep GPU cluster integration
Free tier + paid plans · from $50 USD/mo
US
Founded 2017
Self-hostable
No ratings yet
The wandb Python library instruments training runs capturing metrics, system stats, checkpoints, and media. Sweeps runs Bayesian or grid hyperparameter searches. The W&B API enables programmatic access to run histories and artifacts. Acquired by CoreWeave in March 2025.