CVE-2025-32434
criticalpriority · attend · 50EPSS 0.024 · 84 %ileexploit · PoCDEEURUUSEPSS · scored 2026-10-27
PyTorch is a Python package that provides tensor computation with strong GPU acceleration and deep neural networks built on a tape-based autograd system. In version 2.5.1 and prior, a Remote Command Execution (RCE) vulnerability exists in PyTorch when loading a model using torch.load with weights_only=True. This issue has been patched in version 2.6.0.
cvss
Technical severity, 0–10. One row per scoring source.
base score 0–10 · higher = more severe · hover a vector to decode⚠ sources disagree · high–critical · 8.9–9.8
ssvc
A stakeholder decision — act, attend, or track — from exploitation and impact.
weaknesses
The CWE weakness classes this flaw belongs to.
exploits (4)
Public proof-of-concept and weaponised exploit code.
- AlexanderGumeniuk/CVE-2025-32434AlexanderGumeniuk/CVE-2025-32434PoC · poc_in_github
- cyhe50/cve-2025-32434-poccyhe50/cve-2025-32434-pocPoC · poc_in_github
affected products (14)
Products and version ranges known to be affected.
advisories (4)
Vendor and ecosystem advisories describing this.
- WID-SEC-W-2025-0857PyTorch: Schwachstelle ermöglicht CodeausführungDeutsch2025-04-21csaf · cert_bundDE
Ein entfernter, anonymer Angreifer kann eine Schwachstelle in PyTorch ausnutzen, um beliebigen Programmcode auszuführen.
- GHSA-53q9-r3pm-6pq6PyTorch: `torch.load` with `weights_only=True` leads to remote code execution2025-04-18ghsa · osvUS
# Description I found a Remote Command Execution (RCE) vulnerability in PyTorch. When loading model using torch.load with weights_only=True, it can still achieve RCE. # Background knowledge https://github.com/pytorch/pytorch/security As you can see, the PyTorch official documentation considers using `torch.load()` with `weights_only=True` to be safe.  Since everyone knows that weights_only=False is unsafe, so they will use the weights_only=True to mitigate the seucirty issue. But now, I just proved that even if you use weights_only=True, it can still achieve RCE. # Credit This vulnerability was found by Ji'an Zhou.
- BIT-pytorch-2025-32434PyTorch: `torch.load` with `weights_only=True` leads to remote code execution2025-04-22vendor · osvUS
PyTorch is a Python package that provides tensor computation with strong GPU acceleration and deep neural networks built on a tape-based autograd system. In version 2.5.1 and prior, a Remote Command Execution (RCE) vulnerability exists in PyTorch when loading a model using torch.load with weights_only=True. This issue has been patched in version 2.6.0.
- PYSEC-2025-412025-04-18vendor · osvUS
PyTorch is a Python package that provides tensor computation with strong GPU acceleration and deep neural networks built on a tape-based autograd system. In version 2.5.1 and prior, a Remote Command Execution (RCE) vulnerability exists in PyTorch when loading a model using torch.load with weights_only=True. This issue has been patched in version 2.6.0.
also known as (9)
Other names and identifiers for this entity.
sources
Which sources assert this record, and when.
references (5)
External citations — patches, advisories, analysis.
- https://access.redhat.com/security/cve/cve-2025-32434cnnvd
- https://github.com/pytorch/pytorch/security/advisories/GHSA-53q9-r3pm-6pq6cnnvd
- https://github.com/pytorch/pytorch/security/advisories/GHSA-53q9-r3pm-6pq6enisa_euvd
- https://github.com/pytorch/pytorch/security/advisories/GHSA-53q9-r3pm-6pq6Vendor Advisorynvd
- https://lists.debian.org/debian-lts-announce/2025/12/msg00000.htmlnvd
Attackers weaponise vulnerabilities like this against internet-exposed services. See what is exposed on your perimeter with a free passive scan.
Free scan