Vulnerability Database · PyPI · numpy · Version
numpy 2.4.1
2.4.1 is affected by 2 vulnerabilities.
The vulnerabilities that cover this exact version are listed below with the fixed version to move to. The upgrade Vulert recommends: 2.5.4.
Critical 2 High 0 Medium 0 Low 0 Unrated 0
-
Released 10 Jan 2026
Licence 0BSD AND BSD-3-Clause AND CC0-1.0 AND MIT AND Zlib
Also known from other sources 0
Newer releases 135
Database updated 23/Oct/2026
Release and licence data published by PyPI. Licence information is not legal advice.
Concerned about your application's security?
Vulert continues monitoring your dependencies and alerts you when new vulnerabilities are disclosed. No installation and no source-code access required.
01 Vulnerabilities affecting this version
numpy 2.4.1 vulnerabilities
2 vulnerabilities cover numpy 2.4.1. Open the full list in the database.
| Vulnerability | Severity | Fixed version | Disclosed |
|---|---|---|---|
| CVE-2019-6446: Unsafe Usage of Pickle Module in NumPy | 9.8 critical | No fix published | 24 May 2022 |
| CVE-2019-6446: Numpy Deserialization of Untrusted Data | 9.8 critical | No fix published | 24 May 2022 |
Looking for another version, or everything Vulert holds for numpy?
Questions
About numpy 2.4.1
Is numpy 2.4.1 vulnerable?
Yes. 2 vulnerabilities in the Vulert database cover numpy 2.4.1: 2 critical, 0 high, 0 medium and 0 low. The upgrade Vulert recommends is 2.5.4, and each entry above names the fixed version to move to.
Which version of numpy should I use instead of 2.4.1?
2.5.4 is the version Vulert recommends for this release line. The version table on the package page lists every release with the vulnerabilities recorded against it.
Is 2.4.1 the latest release of numpy?
No. 2.4.1 was published 10 Jan 2026, and newer releases exist. Newer is not automatically safer — the version table on the package page shows what each release carries.
How do I monitor numpy for upcoming vulnerabilities?
Upload the manifest or SBOM for your application at vulert.com/scanner and Vulert keeps checking the numpy version you actually run against new disclosures, then alerts you when one affects it. There is no installation and no source-code access required.