# gradio-app/trackio

A lightweight, local-first, and free experiment tracking library from Hugging Face 🤗

Repository: https://github.com/gradio-app/trackio
Canonical: https://ross.abutalabs.com/products/trackio
Language: Python
License: MIT
License Family: permissive
Last push: 2026-08-26T17:41:05+00:00

## Health v2 (maintenance only)
Score: 82/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 87, longevity 34
- inputs: {"age_days": 482, "days_push": 7, "days_rel": 7, "gap_med": 2.5, "n_releases_24m": 61}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1662, forks 151 (observed 2026-08-28T04:05:18.707944+00:00)

## What it is
Trackio is a lightweight, free, local-first experiment tracking library from Hugging Face that stores metrics in SQLite with an optional Gradio-style dashboard and Hugging Face Space hosting. It is API-compatible with wandb, making it a drop-in replacement, and includes a CLI for SQL-based querying suited to LLM-driven analysis.

## Use cases
- track machine learning experiment metrics locally without an account
- replace wandb with a free drop-in experiment tracker
- let AI agents autonomously run and log ML experiments
- query experiment logs with SQL from the CLI
- share experiment dashboards via a Hugging Face Space

## When to choose
- you want free, local-first experiment tracking with no signup
- you have existing wandb logging code and want a drop-in replacement
- you need programmatic/SQL access to experiment data for LLM analysis

## When to avoid
- you need enterprise features like team access controls or SSO
- you depend on advanced wandb features like sweeps or artifacts beyond basic logging

## Facets
- artifact type: library
- maturity: active
- function: monitoring, data-visualization, cli, analytics
- domain: machine-learning, data-science, developer-tools, artificial-intelligence
- platform: python, cli, cross-platform, self-hosted
- tags: experiment-tracking, mlops, wandb-compatible, sqlite, llm-friendly, hugging-face, metrics-logging

## Member repositories
- gradio-app/trackio (main) score 82

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:18.707944+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:44:34.137636+00:00, confidence not recorded.
  - readme: https://github.com/gradio-app/trackio (fetched 2026-08-28T04:05:18.707944+00:00, sha 98d1e8fcdad5)
  - registry_pypi: https://pypi.org/pypi/trackio/json (fetched 2026-08-29T11:16:45.678890+00:00, sha c3f0e37eb21f)
- Data as of 2026-08-30T08:39:29.467469+00:00.