# visdom

> Real-time visualization and monitoring for AI/ML experiments

Visdom is a visualization and monitoring tool designed for scientific experimentation and machine learning workflows. It broadcasts real-time plots, images, embeddings, and metric dashboards from training loops directly to an interactive browser interface. It runs as a Python server via CLI or Docker and integrates into code through a Python client library.

- Source: https://github.com/fossasia/visdom
- Homepage: https://visdom.dev
- License: Apache-2.0
- Built with: Python 81.6%, JavaScript 14.8%, CSS 2.0%, Lua 1.2%, Other 0.4%
- Stars: 10327
- Forks: 1292
- Contributors: 130
- Last commit: 2026-10-27
- Latest release: v0.3.0 (2026-09-11)
- Purpose: AI & LLM tooling, Data & analytics
- Type: Web, Docker / self-host, CLI, Library / SDK
- For: Personal

## Worth score: 78/100

How much DigGitHub recommends it, from activity, adoption, docs, license and security signals.

- Popularity: 20/25
- Momentum: 10/20
- Maintenance: 25/25
- Community: 11/15
- Readiness: 12/15

## Editor note

Machine Learning

## Downloads

Latest release: [Release 0.3.0](https://github.com/fossasia/visdom/releases/tag/v0.3.0). Files from the project's GitHub release, not checked by DigGitHub.

Source code only: this release has no files to download.

## Alternatives

- [data-formulator](https://diggithub.com/microsoft/data-formulator.md): AI-powered interactive data exploration and visualization
- [Label Studio](https://diggithub.com/HumanSignal/label-studio.md): Data labeling for any data type
- [TrendRadar](https://diggithub.com/sansan0/TrendRadar.md): AI-driven public opinion monitor and multi-platform trend aggregator

---

Source page: https://diggithub.com/fossasia/visdom
Updated: 2026-10-29