PyPI Package Downloads Tracker

PyPI Package Downloads Tracker fetches daily/weekly/monthly download counts, cumulative totals, OS and Python version breakdowns, and package metadata for any PyPI package — from free, no-auth APIs (pypistats.org + PyPI JSON API). Compare multiple packages side-by-side with compareMode. No API key required. $0.002/package.

PyPI download statistics are the most objective signal of Python library adoption. Whether you're choosing between competing libraries, tracking your own package's growth, or auditing dependencies, this actor gives you the numbers in structured form.

---

Why use PyPI Package Downloads Tracker?

1. Library selection — Compare download velocity of competing libraries (numpy vs. polars vs. pandas) before adopting one for production. Monthly downloads + growth trend is the most honest signal of ecosystem momentum.
2. Own package growth tracking — Monitor your published PyPI package's daily/weekly/monthly downloads. Sudden spikes often reveal unexpected users; sudden drops reveal breaking changes.
3. Dependency audit — Before adding a dependency, check whether it's actively maintained and widely used. Very low download counts = potential orphaned library risk.
4. Competitive intelligence — Track downloads for AI libraries (anthropic, openai, transformers, langchain) to understand relative adoption of competing SDKs in real time.
5. OS/Python version breakdown — includeSystemBreakdown: true shows which Python versions and operating systems dominate the user base — critical for setting support matrix priorities.

---

How to use

1. Open the actor on Apify Store and click Try for free.
2. Set packages to the PyPI package names you want to analyze (e.g. ["numpy", "pandas", "polars"]).
3. Enable compareMode: true to get a side-by-side ranking record.
4. Enable includeSystemBreakdown: true for OS + Python version data.
5. Click Start. Returns one structured record per package.

---

Input

Field	Type	Default	Description
packages	string[]	required	PyPI package names to analyze
includeMetadata	boolean	true	Fetch version, author, license, summary from PyPI
includeSystemBreakdown	boolean	false	Per-OS and per-Python-version download breakdown
includeOverall	boolean	true	Cumulative totals with and without mirrors
compareMode	boolean	false	Append a summary record ranking packages by monthly downloads

---

Output

{
  "package": "numpy",
  "downloads_last_day": 1234567,
  "downloads_last_week": 8901234,
  "downloads_last_month": 34567890,
  "downloads_with_mirrors": 35000000,
  "downloads_without_mirrors": 34567890,
  "rank": 1,
  "version": "1.26.4",
  "author": "Travis E. Oliphant et al.",
  "license": "BSD-3-Clause",
  "summary": "Fundamental package for array computing in Python",
  "homepage": "https://numpy.org",
  "pypi_url": "https://pypi.org/project/numpy/",
  "fetched_at": "2026-09-02T12:00:00+00:00"
}

With includeSystemBreakdown: true:

{
  "system_breakdown": {"Linux": 20000000, "Windows": 9000000, "Darwin": 5000000},
  "python_breakdown": {"3.11": 12000000, "3.12": 10000000, "3.10": 8000000}
}

---

Cost estimation

Pay-Per-Event: $0.005 startup + $0.002 per package analyzed.

Use case	Packages	Estimated cost
Single package check	1	~$0.007/run
Compare 5 packages	5	~$0.015/run
Monitor 10 packages weekly	10	~$0.025/week
Compare 20 AI libraries	20	~$0.045/run

---

FAQ

How current is the download data? pypistats.org updates daily, typically with a ~24-hour lag. Very new packages (< 1 day on PyPI) may return null for counts.

What's the difference between with_mirrors and without_mirrors? downloads_with_mirrors includes CDN-served downloads from pip mirrors and CI caches. downloads_without_mirrors is closer to "direct user installs." For most analysis, use downloads_without_mirrors to avoid inflated numbers from automated mirrors.

Does it include GitHub Actions / CI downloads? Yes — pip installs from CI/CD pipelines (GitHub Actions, CircleCI, etc.) are counted in download statistics. This is a significant portion of downloads for popular packages.

Can I track private packages? No — the actor only accesses public PyPI packages via the public API. Private packages hosted on PyPI or private registries (Artifactory, devpi) are not accessible.

---

Related actors

• AI Model Tracker — Track AI provider models alongside SDK (anthropic, openai) download trends
• GitHub Repo Stats Tracker — GitHub stars/forks to complement PyPI download statistics
• PyPI Package Analyzer — Deeper analysis: vulnerability checks, dependency graphs, security audits

Feedback

If this actor helps your library research, a review helps others find it: Leave a review on Apify Store