How severe is CVE-2021-4118?

CVE-2021-4118 has a CVSS score of 7.8/10, rated HIGH. Immediate patching is strongly recommended.

Which packages are affected by CVE-2021-4118?

CVE-2021-4118 affects the following packages: pytorch-lightning (PyPI). Ecosystems affected: PyPI.

How do I fix CVE-2021-4118?

Update pytorch-lightning to 1.6.0 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms CVE-2021-4118 is resolved across your whole dependency graph.

How do I detect CVE-2021-4118 in my PyPI dependencies?

Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for pytorch-lightning, including transitive dependencies — a direct dependency you never call can still pull in a vulnerable version.

How do I mitigate CVE-2021-4118 if there is no patch (or I can't update yet)?

Do not deserialise data from untrusted sources: where the format allows it, restrict deserialisation to an explicit allowlist of expected types, and prefer a data-only format (JSON, Protobuf) over one that can reconstruct arbitrary objects until you can upgrade.

Is CVE-2021-4118 actively exploited in the wild?

Yes. There are 1 known exploit references for CVE-2021-4118, including 1 in-the-wild exploitation observed on GitHub and other sources. Treat this as actively exploitable and prioritize patching immediately. All exploit code should only be run in an isolated sandbox environment for research or authorized testing — never against production systems without explicit written authorization.

What is the EPSS score for CVE-2021-4118?

CVE-2021-4118 has an EPSS (Exploit Prediction Scoring System) score of 1.0%, placing it in the 64th percentile of all CVEs. EPSS is maintained by FIRST.org and estimates the probability that a vulnerability will be exploited in the wild within the next 30 days. This score indicates relatively lower exploitation probability, though the CVSS severity should still guide your patching priority.

What type of vulnerability is CVE-2021-4118?

CVE-2021-4118 is classified as Deserialization of Untrusted Data (CWE-502). This weakness type describe the underlying flaw category, which helps determine the potential impact and the right class of mitigation. This is a high-impact weakness class that often enables remote code execution or data exposure.

When was CVE-2021-4118 published, and has it been updated?

CVE-2021-4118 was published on December 23, 2021 and was last updated on October 24, 2026. Advisory data evolves as severity scores, affected ranges, and exploit intelligence are revised — always check the latest version of the advisory before acting.

🐍

🐍 PyPI

Not in CISA KEV

HIGH severity

CVE-2021-4118 — pytorch-lightning

HIGHFix: pytorchlightning/pytorch-lightning@62f1e82

CVE-2021-4118 is a high-severity (CVSS 7.8) Deserialization of Untrusted Data vulnerability in pytorch-lightning. 1 public exploit or proof-of-concept reference is indexed, so weaponization risk is real. A fix is available for pytorch-lightning — see the affected versions and patch details below.

pytorch-lightning is vulnerable to Deserialization of Untrusted Data

Also known asGHSA-2vj5-px25-gjrpPYSEC-2021-874PYSEC-2026-3968

Published

Dec 23, 2021

Updated

Oct 24, 2026

Affected

1 pkg

Patched

1 / 1

Exploits

1 known

Exploitation data as of Nov 12, 2026 · OSV.dev, NVD, FIRST.org (EPSS)

EPSS Exploitation Probability

via FIRST.org ↗

1.0%probability of exploitation in next 30 days

Lower Risk+0.01%

Lower risk than most CVEs64th percentile — riskier than 64% of all scored CVEsHighest risk

Probability of exploitation in the next 30 days, from FIRST.org EPSS.

How urgent is this, really

Where this sits among everything scored

Real-World Exposure

1 pkg affected

🐍pytorch-lightning

Real-time download stats are indexed for npm and PyPI packages. This vulnerability affects PyPI packages — download data is not available via public APIs for these ecosystems.

Description

pytorch-lightning is vulnerable to Deserialization of Untrusted Data

Affected Packages

1 total 1 fixed

Ecosystem	Package	Vulnerable range	Fix
🐍PyPI	`pytorch-lightning`	all versions	1.6.0`pip install --upgrade 'pytorch-lightning==1.6.0'`

Affected Products

1 product · 1 configurations

Application

pytorch lightninglightningai

< 1.6.0

range

Exploits & PoCs

Research use only. For defensive security, authorized penetration testing, and academic research only. Never execute exploit code against systems without explicit written authorization.

pytorch-lightning is vulnerable to Deserialization of Untrusted Data

huntr.devNVDJan 2022

Detection & mitigation playbook

Open-source dependency

Detect
Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for pytorch-lightning, including transitive dependencies — a direct dependency you never call can still pull in a vulnerable version.
Fix
Update pytorch-lightning to 1.6.0 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms CVE-2021-4118 is resolved across your whole dependency graph.
Workarounds
Do not deserialise data from untrusted sources: where the format allows it, restrict deserialisation to an explicit allowlist of expected types, and prefer a data-only format (JSON, Protobuf) over one that can reconstruct arbitrary objects until you can upgrade.

Frequently Asked Questions

pytorch-lightning is vulnerable to Deserialization of Untrusted Data

O3 Security · Impact-Aware SCA

Is CVE-2021-4118 in your dependencies?

Find it across PyPI, including transitive dependencies.

Scan my dependencies How O3 SCA works

CVSS Score

7.8/10CVSS v3 base

Vector breakdown

Attack Vector

Local

Attack Complexity

Low

Privileges Required

None

User Interaction

Required

Scope

Unchanged

Confidentiality

High

Integrity

High

Availability

High

Raw vector

CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H

CWE Classification

CWE-502Deserialization of Untrusted Data

Weaponization

No Nuclei template or Metasploit module yet. 1 public proof-of-concept reference is listed on this page.

OpenSSF Scorecard

Package health score

pytorch-lightning

6.4/10

Code-Review10/10

Maintained10/10

Vulnerabilities10/10

Dependency-Update-Tool10/10

Branch-Protection1/10

Assessed Aug 15, 2022

Timeline

Disclosed

Dec 23, 2021

First exploit

Jan 4, 2022

Last updated

Oct 24, 2026

Identifiers

CVE-2021-4118

GHSA-2vj5-px25-gjrp

PYSEC-2021-874

PYSEC-2026-3968

References

Advisorynvd.nist.gov · CVE 2021 4118 Advisorygithub.com · GHSA 2vj5 px25 gjrp Fixgithub.com · 62f1e82e032eb16565e676d39e0db0cac7e34ace Fixhuntr.dev · 31832f0c e5bb 4552 a12c 542f81f111e6 Webgithub.com Webgithub.com Webgithub.com · 1.6.0 Webgithub.com · PYSEC 2021 874.yaml Packagegithub.com · pytorch lightning

Related vulnerabilities

Also affects pytorch-lightning

This CVE is high severity; 5 of 5 other CVEs affecting pytorch-lightning are rated as high or higher (top: critical).

CVE-2026-31221CVE-2026-44484CVE-2024-8019CVE-2024-8020CVE-2022-0845

Same weakness: CWE-502

This CVE is high severity; 2 of 3 other CVEs with CWE-502 are rated as high or higher (top: critical).

CVE-2026-128083CVE-2026-114151CVE-2026-126187Browse all CWE-502

Provenance

Sources

OSV.devNVDFIRST.org (EPSS)

Last verifiedOct 24, 2026

CVE-2021-4118 — pytorch-lightning

EPSS Exploitation Probability

How urgent is this, really

Where this sits among everything scored

Real-World Exposure

Description

Affected Packages

Affected Products

Detection & mitigation playbook

Detect

Fix

Workarounds

Frequently Asked Questions

Is CVE-2021-4118 in your dependencies?