# Weights & Biases — ML Experiment Tracking

> W&B tracks, visualizes, and manages ML experiments and LLM apps. 10.9K+ GitHub stars. Experiment tracking, model versioning, Weave for LLMs. MIT.

## Install

Save in your project root:

## Quick Use

```bash
pip install wandb
wandb login

python -c "
import wandb
wandb.init(project=\"my-project\")
wandb.log({\"accuracy\": 0.95, \"loss\": 0.05})
wandb.finish()
print(\"Logged to W&B!\")
"
```

---

## Intro

Weights & Biases (W&B) is the AI developer platform for tracking, visualizing, and managing machine learning experiments. With 10,900+ GitHub stars and MIT license, it provides experiment tracking and visualization, hyperparameter management, model and data versioning, LLM app monitoring via Weave, and integration with TensorFlow, PyTorch, Keras, JAX. Cloud, dedicated, or self-managed hosting.

**Best for**: ML teams needing experiment tracking and model management
**Works with**: Claude Code, OpenAI Codex, Cursor, Gemini CLI, Windsurf

---

## Key Features

- Experiment tracking and visualization
- Hyperparameter sweeps
- Model and data versioning
- Weave for LLM observability
- PyTorch, TensorFlow, Keras, JAX integration
- Cloud or self-hosted

---

### FAQ

**Q: What is W&B?**
A: ML experiment tracking platform with 10.9K+ stars. Tracking, visualization, versioning, LLM monitoring via Weave. MIT.

**Q: How do I install W&B?**
A: pip install wandb. wandb login. wandb.init() in your code.

---

## Source & Thanks

> [wandb/wandb](https://github.com/wandb/wandb) — 10,900+ GitHub stars

---
Source: https://tokrepo.com/en/workflows/084bc7ee-ad89-498d-8640-9cbd2e197a42
Author: AI Open Source