Advisory record · GHSA-4f3f-g24h-fr8m

Keras has an untrusted deserialization vulnerability

CVSS 8.8 HighEPSS 0.0042 · 36th percentile

Published by the GitHub Advisory Database at 13 Apr 2026, 15:31 UTC. agentprime has not analysed this yet, so whether it affects any app is not known.

PackageEcosystemAffectedFixed in
kerasPyPI<3.13.23.13.2
Identifiers
GHSA-4f3f-g24h-fr8m · CVE-2026-1462 · PYSEC-2026-2547
Weakness (CWE)
CWE-502
Published
13 Apr 2026
Updated
10 Sept 2026

Machine-written summary

Summary written by Claude Haiku from the source advisory. Not reviewed by a person.

Keras (PyPI) versions before 3.13.2 are affected; the source describes the flaw in version 3.13.0 in the TFSMLayer class. An attacker who can supply a crafted .keras model file can get a TensorFlow SavedModel loaded even with safe_mode=True, which can lead to arbitrary code execution under the victim's privileges during model inference. Upgrade to keras 3.13.2 or later.

Written 1 Nov 2026.

Not analysed yet

We haven’t published our analysis of this Advisory yet. Once a person has reviewed it, this page will say what’s vulnerable, what has to be true for your app to be affected, and how to check.

Severity and exploitation

ScoreRatingCVSSVectorSource
8.8 High 3.0 CVSS:3.0/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H CNA (@huntr_ai)
  • CISA KEV Not listed.
  • EPSS 0.0042, the 36th percentile, on 12 Nov 2026: FIRST’s estimate of the chance of exploitation in the next 30 days.

Timeline

  1. 31 Oct 2026, 14:11 UTC First seen by agentprime
  2. 31 Oct 2026, 14:11 UTC Confirmed by OSV

Sources

Facts on this page come from the GitHub Advisory Database (CC-BY 4.0), OSV and the other sources listed, as published. We don’t reproduce full advisory descriptions, which can contain exploit code.