Weights & Biases
MLOps platform for experiment tracking, model registry, and LLM observability.
Acquired — still operating
acquired by CoreWeave May 2025 (~$1.4B)
About
Weights & Biases (W&B) is an AI developer platform covering the full lifecycle of model and LLM application development. W&B Models provides experiment tracking, hyperparameter sweeps, artifacts, model registry, and reports, while W&B Weave adds LLM tracing, evaluation, cost tracking, and guardrails for production AI applications.
In May 2025, CoreWeave acquired Weights & Biases for roughly $1.4 billion, folding it into an end-to-end AI cloud story alongside GPU infrastructure. In 2026 CoreWeave launched Forge, a development layer combining W&B Models with inference, training, and agent tooling. W&B remains usable with any infrastructure provider, not just CoreWeave. Limitation: the acquisition means the product roadmap now follows CoreWeave's platform strategy, and heavy users should watch how Forge packaging affects standalone W&B pricing.
What it can do
- Experiment tracking and visualization
- Hyperparameter sweeps
- Model registry and artifacts
- LLM tracing and evaluation via Weave
Setup
Sign up at wandb.ai; install the Python SDK with pip install wandb and log in via wandb login.
Prerequisites
A Python environment for the SDK; a W&B account for hosted tracking.
Use cases
Works with
Trust score — 9/10
Scored 9/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 63; Trending score 88/100 from researcher assessment.
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.