CVE-2025-3136 PyTorch CUDACachingAllocator.cpp torch.cuda.memory.caching_allocator_delete memory corruption

Medium 4.8 Published Updated

Local attackers can exploit memory corruption in PyTorch's CUDA caching allocator to potentially execute arbitrary code or cause denial of service.

Exploited in the wild
Not known
Login needed
Any signed-in user
Attackable
Local or same network only
Victim action needed
Not needed
Fix
Available

What is CVE-2025-3136?

CVE-2025-3136 is a medium-severity memory corruption vulnerability in PyTorch, published on 3 Apr 2025. It can only be exploited from the same machine or local network, with no action from the victim.

Official description

A vulnerability, which was classified as problematic, has been found in PyTorch 2.6.0. This issue affects the function torch.cuda.memory.caching_allocator_delete of the file c10/cuda/CUDACachingAllocator.cpp. The manipulation leads to memory corruption. An attack has to be approached locally. The exploit has been disclosed to the public and may be used.

Weakness: CWE-787, CWE-119

Which versions are affected by CVE-2025-3136?

2 open source packages ship an affected version, including pytorch and torch. It affects 1 product.

PackageAffectedFixed in
pytorch BitnamiFrom 2.6.02.7.0
torch PyPIAll versions to 2.6.0-NANone

Products include Linuxfoundation Pytorch.

How do I fix CVE-2025-3136?

Upgrade pytorch to 2.7.0 or later.

Upgrade to the fixed version. Memory-safe languages, compiler hardening and fuzzing prevent most new cases. More about Memory corruption.

Is CVE-2025-3136 being exploited?

It is not confirmed as exploited in the wild, but public exploit code for CVE-2025-3136 exists.

Sources

Also known as BIT-pytorch-2025-3136, DEBIAN-CVE-2025-3136, ECHO-ff69-5060-f4ce, GHSA-r37f-v2p6-x6cx, PYSEC-2025-197, UBUNTU-CVE-2025-3136.

Is CVE-2025-3136 in your software?

Upload an SBOM and sbomify checks it against every vulnerability here, every day.

Check your SBOMs for free