PytorchApplication · Linuxfoundation
CVE-2025-32434
Official description Straight from the sourceThe vendor's or NVD's own wording, published unedited. Authoritative, but often terse — it says what broke, rarely what to do.
NVD · uneditedPyTorch 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.
Technical summary Written by usOur analysis, written from the advisory, the CVSS vector and the affected-version data. It adds context the advisory leaves out, and never invents facts that are not in the source.
dbcve analysis · high confidencePyTorch versions 2.5.1 and prior contain a Remote Command Execution vulnerability that can be triggered when loading models using torch.load with weights_only=True. Despite weights_only being intended to safely load only tensor weights without executing arbitrary Python objects, an attacker can craft a malicious pickled model file to achieve arbitrary code execution.
Verify against the referenced sources before acting — the references below are authoritative for this CVE, this summary is not.
Affected products & versions What the vendor confirmedThe version ranges the vendor confirmed as vulnerable. If your version sits inside a range here, treat yourself as exposed until you have upgraded.
NVD · CPE data
PytorchApplication
< 2.6.0CVSS breakdown How the score is builtThe industry scoring standard. It rates how the flaw is reached, what it takes to exploit, and what an attacker gains — the score is derived from those, not the other way round.
From the vector- Attack vector
- Network
- Complexity
- Low
- Privileges
- None
- User interaction
- None
- Scope
- Unchanged
- Confidentiality
- High
- Integrity
- High
- Availability
- High
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
Am I affected? How to checkSteps we derive from the advisory and the affected-version data, so you can decide whether this CVE reaches your setup. They are a guide, not a scan — your own configuration is the authority.
dbcve checksWork through these to decide whether this CVE applies to you.
-
Identify installed PyTorch versionRun 'python -c "import torch; print(torch.__version__)"' or check your dependency lock file (requirements.txt, setup.py, pyproject.toml, pip freeze output)Affected if Version is 2.5.1 or lower, or any version below 2.6.0
-
Find torch.load usage in codeSearch source code repositories for patterns like 'torch.load(' or 'torch.load(' with grep, IDE search, or code reviewAffected if Codebase contains torch.load calls loading untrusted model files
-
Check if weights_only=True is usedSearch for 'weights_only=True' in code using grep or IDE search, or review model loading scriptsAffected if weights_only=True is set - this is the specific configuration that enables the attack surface described in the CVE
-
Identify source of loaded model filesReview where model files loaded via torch.load originate - check if they come from untrusted sources, user uploads, external URLs, or unverified directoriesAffected if Models are loaded from untrusted or user-controlled locations without verification
-
Audit model loading pipelinesExamine any ML pipelines, inference scripts, model serving code, or data processing jobs that use torch.load, particularly in production or shared environmentsAffected if Production systems or shared environments load pickled model files without security verification
You are affected if your installed PyTorch version is below 2.6.0 and your code uses torch.load to load pickled model files from untrusted sources, even with weights_only=True set.
Generated from the published advisory. Verify against your own configuration.
Remediation Closing itWhat it takes to close this. Where a vendor fix exists we point at it; where none exists we say so plainly, and can build one. Effort estimates are scoped from the advisory, not from your codebase.
dbcve · scoped2.6.0
Upgrade PyTorch to version 2.6.0 or later to patch the vulnerability. Review all code paths that use torch.load and ensure weights_only=True is properly set where applicable.
PyTorch 2.6.0
- Upgrade PyTorch to version 2.6.0 or later by running: pip install --upgrade torch (or conda update pytorch if using conda)
- Verify the installation succeeded by running: python -c 'import torch; print(torch.__version__)'
- Ensure the version printed is 2.6.0 or higher
Generated from the published advisory — verify against the referenced sources before acting.
Fix this in Pytorch
Scoped from the published advisory
- Consultation4.0 h
- Implementation4.0 h
- Testing8.0 h
- Review / QA4.0 h
An estimate, not a bill — we confirm scope with you before any work starts. Need it this week? Rush from $5,504.
Scan for this in your stack
Free · runs locallyCheck whether your project pulls in CVE-2025-32434 — or any other known-vulnerable package — straight from your lock files. Free and open source; it runs locally and uploads nothing.
References Go to the primary sourcePrimary sources — vendor advisories, patches and trackers. Where our summary and a reference disagree, the reference wins.
Primary sourcesPractitioner notes
ContributedPeer-ranked notes from engineers who’ve handled CVE-2025-32434 in production — separate from our analysis above.
The advisory tells you what broke. It rarely tells you what actually worked. If you’ve dealt with this one, that detail is what the next engineer is searching for.
- The version that genuinely resolved it — not the one the vendor claimed
- A config change or rule that shut the vector down
- A gotcha in the upgrade path that cost you an afternoon
No notes yet
Be the first to add a field note for this CVE — a mitigation you’ve verified, a version caveat, or a link to a working fix. Sign in above to contribute.
A place for practitioners to share what actually worked: a mitigation you’ve tested, a configuration change, a version- or environment-specific caveat, or a link to a verified patch. The most useful notes rise to the top as peers upvote them, so the signal stays high.
- Verified mitigations, workarounds, and config changes
- Version or environment caveats, and links to real fixes
- No weaponised exploit code, or anything meant to cause harm
- No spam, self-promotion, credentials, or personal data