#4444th of 75 in ML frameworks and libraries
wandb
Python SDK for tracking ML experiments and versioning datasets and models
Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 11.3k
- License
- MIT
- Last commit
- Oct 2026
- Last release
- Oct 2026
- Language
- Python
10,856 on 31 Mar 202611,305 on 20 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
The wandb Python client logs hyperparameters and metrics from training scripts or notebooks to a W&B server, where runs are visualized and compared. It also covers dataset and model versioning, with integrations for common ML frameworks. The backend is a hosted service or a W&B Server install on AWS, GCP, Azure or on-premises.
Who it is for: ML engineers tracking training experiments and datasets
Strengths
- Adds tracking with wandb.init() and run.log() in an existing training script
- Integrates with popular ML frameworks and libraries for tracking and data versioning
- Hosted multi-tenant, dedicated cloud, or self-managed server options
- Weave, a separate suite, covers tracing and evaluation of LLM apps
Weaknesses
- This repo is the client SDK; the server and dashboard are not included
- Needs a W&B account and API key for the default hosted setup
- Self-managed server requirements and licensing are not described in the README
- LLM tracing lives in Weave, a separate project
What it needs
- no GPU
- Needs W&B account and API key, W&B Cloud or W&B Server
Also in ML frameworks and libraries
See all 75| Rank | Project | Score |
|---|---|---|
| 1 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 2 | pytorchPython tensor library with GPU support and tape-based autograd for deep learning | 94 out of 100 |
| 3 | ultralyticsPython library and CLI for training and running YOLO vision models | 92 out of 100 |
| #4 | pytorch-image-modelsPyTorch library of image models, pretrained weights and reference training scripts | 91 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 90 out of 100 |