CVE-2021-29596: TensorFlow GPU - Division by Zero Error in EmbeddingLookup Operator
Summary
CVE-2021-29596: TensorFlow is an end-to-end open source platform for machine learning. Affects tensorflow-gpu across 5 version ranges.
< 2.1.4
>= 2.2.0, < 2.2.3
>= 2.3.0, < 2.3.3
>= 2.4.0, < 2.4.2
< f61c57bd425878be108ec787f4d96390579fb83e
Fixed in 4 releases; the minimum fixed version depends on the release line you run. How to fix ↓
Also tracked for
- tensorflow · Bitnami
- tensorflow · PyPI
- tensorflow · PyPI
- tensorflow-cpu · PyPI
The same CVE-2021-29596 is tracked separately for each ecosystem and vendor. Its fix information is the same; the package and affected versions differ.
Check if this vulnerability affects your application
Upload a manifest or SBOM and we'll show whether you're affected, plus every other risky package in the same scan.
No signup required · Results in about 60 seconds
Background
TensorFlow is an end-to-end open source platform for machine learning. It provides a comprehensive ecosystem for building and deploying machine learning models. The TensorFlow GPU package, available on PyPI, is specifically designed to leverage the power of GPUs for accelerated machine learning computations. It is widely used in various industries and research domains for developing and training deep learning models.
Vulnerability Detail
The vulnerability (CVE-2021-29596) affects the TensorFlow GPU package. Specifically, it resides in the implementation of the EmbeddingLookup TFLite operator. The vulnerability arises due to a division by zero error in the code. An attacker can exploit this vulnerability by crafting a malicious model in which the first dimension of the value input is set to 0. When this model is executed, it triggers the division by zero error, leading to a potential crash or undefined behavior. The vulnerability has been identified and fixed in TensorFlow 2.5.0. However, earlier versions, including TensorFlow 2.4.2, TensorFlow 2.3.3, TensorFlow 2.2.3, and TensorFlow 2.1.4, are also affected.
How To Fix
Latest published version: 2.12.0 as of 22 Oct 2026
pip install --upgrade "tensorflow-gpu==2.12.0"Upgrading to it may include breaking changes; the versions below are the minimum fixed release for this vulnerability.
If you can't move to the latest
The minimum fixed version depends on the release line you run.
-
< 2.1.42.1.4pip install --upgrade "tensorflow-gpu==2.1.4" -
>= 2.2.0, < 2.2.32.2.3pip install --upgrade "tensorflow-gpu==2.2.3" -
>= 2.3.0, < 2.3.32.3.3pip install --upgrade "tensorflow-gpu==2.3.3" -
>= 2.4.0, < 2.4.22.4.2pip install --upgrade "tensorflow-gpu==2.4.2"
Tip — Not sure which release line you're on? Vulert can find the fix for the version you are using. Scan free →
Workaround
As a temporary workaround, users can ensure that the first dimension of the value input in their models is not set to 0. This can be achieved by validating the input data and applying appropriate checks before executing the model.
Conclusion
The vulnerability (CVE-2021-29596) in TensorFlow GPU's EmbeddingLookup TFLite operator can be exploited by an attacker through a crafted model with a zero-valued first dimension. This can lead to a division by zero error and potential crashes or undefined behavior. Users are advised to update their TensorFlow GPU package to version 2.5.0 or later to fix the vulnerability. As a temporary workaround, users can ensure that the first dimension of the value input in their models is not set to 0. For further assistance or support in vulnerability remediation, please contact DevNack.com.
See every vulnerability recorded for tensorflow-gpu →
FAQ
TensorFlow GPU is a package that enables GPU acceleration for TensorFlow, allowing for faster training and inference of machine learning models.
The EmbeddingLookup TFLite operator is a component in TensorFlow Lite, a lightweight version of TensorFlow designed for mobile and embedded devices. It is used for performing embedding lookups in neural network models.
An attacker can exploit this vulnerability by crafting a malicious model in which the first dimension of the value input is set to 0. When this model is executed, it triggers a division by zero error, potentially leading to a crash or undefined behavior.
The vulnerability affects TensorFlow GPU versions 2.1.4, 2.2.3, 2.3.3, and 2.4.2. These versions are still within the supported range.
Yes, the vulnerability has been fixed in TensorFlow 2.5.0. It is recommended to update to this version or later to mitigate the risk.
References
Related vulnerabilities
Recent vulnerabilities recorded for tensorflow-gpu.
