CVE-2026-24227
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 · uneditedNVIDIA TensorRT for contains a vulnerability where a user might cause a deserialization of untrusted data. A successful exploit of this vulnerability might lead to code execution.
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 · moderate confidenceNVIDIA TensorRT contains a deserialization vulnerability where processing untrusted data can lead to arbitrary code execution. This is a critical flaw in the model's deserialization logic that allows attackers to craft malicious serialized data that triggers code execution during the deserialization process.
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< 11.0CVSS 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.
-
Check installed TensorRT versionRun 'python -c "import tensorrt; print(tensorrt.__version__)"' or check via 'dpkg -l | grep tensorrt' or 'pip show tensorrt' depending on installation methodAffected if The version displayed is below 11.0 (e.g., 10.x, 9.x, etc.)
-
Identify TensorRT model deserialization usageSearch codebase for TensorRT runtime APIs: IHostMemory, ICudaEngine, or nvinfer1::IRuntime::deserializeCudaEngine calls loading .plan, .onnx, or serialized model filesAffected if Code loads or deserializes TensorRT engine files (.plan) or serialized models from any source
-
Check if TensorRT processes untrusted inputAudit data flow: determine whether model files come from user uploads, network sources, external APIs, or untrusted filesystem locationsAffected if TensorRT processes models from untrusted or external sources without validation
-
Verify TensorRT service exposureReview service configuration: check if TensorRT inference runs as a web service, API endpoint, or networked application accepting model uploadsAffected if TensorRT is exposed via HTTP/API endpoints or accepts remote model input
You are affected if TensorRT version is below 11.0 AND the system deserializes or processes models from untrusted or external sources.
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 · scoped11.0
Apply NVIDIA's official patch when available. Until then, restrict TensorRT model processing to only trusted, verified sources and implement input validation on all deserialized data.
TensorRT 11.0 or later
- 1. Identify the current TensorRT version by running 'python -c "import tensorrt; print(tensorrt.__version__)"' or checking the installed package version
- 2. Download TensorRT version 11.0 or later from the official NVIDIA TensorRT download page (developer.nvidia.com/tensorrt)
- 3. Uninstall the current TensorRT version: 'pip uninstall tensorrt' or use the appropriate package manager for your installation method
- 4. Install the new TensorRT 11.0+ package: 'pip install tensorrt' or follow NVIDIA's installation guide for your platform (ensure you download the correct version for your CUDA runtime)
- 5. Verify the installation succeeded by running 'python -c "import tensorrt; print(tensorrt.__version__)"' and confirming it shows version 11.0 or higher
- 6. Test that your models and inference pipelines function correctly with the new version
- 7. If using containers, rebuild Docker images with the updated TensorRT version
Generated from the published advisory — verify against the referenced sources before acting.
- Consultation8.0 h
- Implementation4.0 h
- Testing16.0 h
- Review / QA8.0 h
An estimate, not a bill — we confirm scope with you before any work starts. Need it this week? Rush from $9,856.
Scan for this in your stack
Free · runs locallyCheck whether your project pulls in CVE-2026-24227 — 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-2026-24227 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