Out-of-bounds Heap Read Vulnerability in PyTorch torchvision Software
CVE-2026-65918

7.1HIGH

Key Information:

Vendor

Pytorch

Status
Vision
Vendor
CVE Published:
23 July 2026

Badges

👾 Exploit Exists🟡 Public PoC

What is CVE-2026-65918?

The torchvision component of PyTorch is vulnerable to an out-of-bounds heap read in the GIF decoder's read_from_tensor callback. The issue arises when the decoder processes malicious or corrupted GIF files, potentially leading to a denial of service through segmentation faults or the exposure of sensitive adjacent heap memory contents. Users are encouraged to upgrade to the fixed version to mitigate this security risk.

Affected Version(s)

vision 0 <= 0.28.0

vision 4e05dc22f5f050a9528cc0ea09ceca6cdaf8f4ed

Exploit Proof of Concept (PoC)

PoC code is written by security researchers to demonstrate the vulnerability can be exploited. PoC code is also a key component for weaponization which could lead to ransomware.

CVE-2026-65918 - Proof of Concept

References

https://github.com/pytorch/vision/issues/9551
technical-descriptionexploit
https://github.com/pytorch/vision/pull/9520
issue-tracking
https://github.com/pytorch/vision/commit/4e05dc22f5f050a9...
patch

CVSS V4

Score:
7.1
Severity:
HIGH
Confidentiality:
Low
Integrity:
None
Availability:
High
Attack Vector:
Network
Attack Complexity:
Low
Attack Required:
None
Privileges Required:
Undefined
User Interaction:
Unknown

Timeline

  • 🟡

    Public PoC available

  • 👾

    Exploit known to exist

  • Vulnerability published

  • Vulnerability Reserved

Credit

George Chen
.