CVE-2026-65918: PyTorch torchvision through 0.28.0, fixed in commit 4e05dc2, contains an out-of-bounds heap read vulnerability

This month

HIGH 7.1WATCHLISTUPGRADE

CVE-2026-65918 is a high-severity flaw (CVSS 7.1) affecting torchvision. Below: why it ranks where it does, which versions are affected, what to do, and the exact commands.

PyTorch torchvision through 0.28.0, fixed in commit 4e05dc2, contains an out-of-bounds heap read vulnerability in the GIF decoder's read_from_tensor callback that passes unclamped length to memcpy. Attackers can supply malicious or truncated GIF files to cause denial of service via segmentation fault or disclose adjacent heap memory contents.

Why this priority

High severity
Public exploit reference
On your watchlist

A fixed release or distro patch exists.

Risk facts

CVSS score	7.1 (High)
Attack vector	Network (remote)
Attack complexity	Low
Privileges needed	None
User interaction	Required
EPSS exploit chance	0.4% (top 62%)
CISA exploited list	No
Weakness	CWE-125
Published	2026-07-23
Last modified	2026-10-24

Publisher (CNA): VulnCheck · Vendor: linuxfoundation

Last updated 2026-10-24 · Sources: CVE Program, NVD, CISA KEV, EPSS and vendor advisories · How we score

Am I affected?

Timeline

2026-07-23 18:17 UTCPublished CVE record published
2026-10-24 13:10 UTCUpdated Record last modified

linuxfoundation torchvision UPGRADE

Affected: <= 0.28.0

Fixed in: newer than 0.28.0

Safe to downgrade to: no unaffected older release

1. Check installed version

pip show torchvision

2. UPGRADE to newer than 0.28.0

pip install --upgrade "torchvision>0.28.0"

Exact fixes by distribution

Fedora 46 · gnome-shell-extension-tiling-assistant · FIXED · 56-1.fc46 · FEDORA-2026-ebf66a5321 · source

sudo dnf upgrade --advisory=FEDORA-2026-ebf66a5321

Debian 12 · pytorch-vision · NO FIX YET

Debian 13 · pytorch-vision · NO FIX YET

RHEL AI (RHEL AI) 3 · rhelai3/bootc-aws-cuda-rhel9 · NO FIX YET

RHEL AI (RHEL AI) 3 · rhelai3/bootc-azure-cuda-rhel9 · NO FIX YET

RHEL AI (RHEL AI) 3 · rhelai3/bootc-azure-rocm-rhel9 · NO FIX YET

RHEL AI (RHEL AI) 3 · rhelai3/bootc-cuda-rhel9 · NO FIX YET

RHEL AI (RHEL AI) 3 · rhelai3/bootc-gaudi-rhel9 · DEFERRED

Public exploits and detection

Public code exists for this CVE. Treat unpatched systems as at risk, and only run PoCs in a lab.

Write-up: github.com/pytorch/vision/issues/9551

Verify at the official source NVD · CVE Program · Report a problem with this page
Scores and fixes here are computed automatically from those and other public sources. Confirm with the vendor advisory before changing production systems.

Quick answers

How do I fix CVE-2026-65918?

torchvision: upgrade to newer than 0.28.0. Exact commands for each OS are on this page.

Is CVE-2026-65918 being exploited?

CVE-2026-65918 is not on the CISA Known Exploited Vulnerabilities list at the time of the last update. Check again, as this can change.

What does CVE-2026-65918 affect?

CVE-2026-65918 affects torchvision. Severity is high (CVSS 7.1). PyTorch torchvision through 0.28.0, fixed in commit 4e05dc2, contains an out-of-bounds heap read vulnerability in the GIF decoder's read_from_tensor callback that passes unclamped length to memcpy. Attackers can supply malicious or truncate

How urgent is CVE-2026-65918?

Priority 44/100. High severity; Public exploit reference; On your watchlist

Software pages

PyTorch vulnerabilities

References

← Back to latest CVEs