PT-2025-39380

Name of the Vulnerable Software and Affected Versions PyTorch versions prior to 2.7.0 Description When inductor is used, nn.Fold experiences an assertion error. Recommendations Update to version 2.7.0 or later...

CVE-2025-46153

PyTorch before 3.7.0 has a bernoullip decompose function in decompositions.py even though it lacks full consistency with the eager CPU implementation, negatively affecting nn.Dropout1d, nn.Dropout2d, and nn.Dropout3d for fallbackrandom=True...

PyTorch 安全漏洞

PyTorch is a Python package open-sourced by PyTorch. PyTorch suffers from an information disclosure vulnerability that stems from FractionalMaxPool2d producing inconsistent results when using torch.compile, no details of the vulnerability are provided at this time...

CVE-2025-55552

pytorch v2.8.0 was discovered to display unexpected behavior when the components torch.rot90 and torch.randnlike are used together...

CVE-2025-55558

A buffer overflow occurs in pytorch v2.7.0 when a PyTorch model consists of torch.nn.Conv2d, torch.nn.functional.hardshrink, and torch.Tensor.view-torch.mv and is compiled by Inductor, leading to a Denial of Service DoS...

CVE-2025-55560

An issue in pytorch v2.7.0 can lead to a Denial of Service DoS when a PyTorch model consists of torch.Tensor.tosparse and torch.Tensor.todense and is compiled by Inductor...

CVE-2025-46152

The PyTorch machine learning framework is vulnerable in versions prior to 2.7.0 (specifically noted as less than 2.2.2-9 in some records) due to an issue in the torch.bitwise_right_shift interface. The vulnerability arises because the function mishandles out-of-bounds values for the **

CVE-2025-55553

A syntax error in the component proxytensor.py of pytorch v2.7.0 allows attackers to cause a Denial of Service DoS...

CVE-2025-55558: Uncontrolled Resource Consumption

A buffer overflow occurs in pytorch v2.7.0 when a PyTorch model consists of torch.nn.Conv2d, torch.nn.functional.hardshrink, and torch.Tensor.view-torch.mv and is compiled by Inductor, leading to a Denial of Service DoS...

CVE-2025-46153: Inefficient CPU Computation

PyTorch before 3.7.0 has a bernoullip decompose function in decompositions.py even though it lacks full consistency with the eager CPU implementation, negatively affecting nn.Dropout1d, nn.Dropout2d, and nn.Dropout3d for fallbackrandom=True...

CVE-2025-46153

PyTorch before 3.7.0 has a bernoullip decompose function in decompositions.py even though it lacks full consistency with the eager CPU implementation, negatively affecting nn.Dropout1d, nn.Dropout2d, and nn.Dropout3d for fallbackrandom=True...

CVE-2025-46153

PyTorch before 3.7.0 has a bernoullip decompose function in decompositions.py even though it lacks full consistency with the eager CPU implementation, negatively affecting nn.Dropout1d, nn.Dropout2d, and nn.Dropout3d for fallbackrandom=True...

CVE-2025-46152

In PyTorch before 2.7.0, bitwiserightshift produces incorrect output for certain out-of-bounds values of the "other" argument...

CVE-2025-46150

In PyTorch before 2.7.0, when torch.compile is used, FractionalMaxPool2d has inconsistent results...

CVE-2025-46148

In PyTorch through 2.6.0, when eager is used, nn.PairwiseDistancep=2 produces incorrect results...

CVE-2025-55557

A Name Error occurs in pytorch v2.7.0 when a PyTorch model consists of torch.cummin and is compiled by Inductor, leading to a Denial of Service DoS...

CVE-2025-46149

The PyTorch machine learning framework contains a vulnerability in the nn.Fold module when inductor is used. Specifically, an assertion error occurs in versions prior to 2.7.0 . While some reports suggest this may be related to buffer overflow attacks, the primary consequence is a denial-of-servi...

CVE-2025-46152: Out-of-bounds Write

In PyTorch before 2.7.0, bitwiserightshift produces incorrect output for certain out-of-bounds values of the "other" argument...

PT-2025-39381

Name of the Vulnerable Software and Affected Versions PyTorch versions prior to 2.7.0 Description When torch.compile is used, the FractionalMaxPool2d function exhibits inconsistent results. This issue affects the accuracy of computations involving this specific function within the PyTorch...

CVE-2025-55552

CVE-2025-55552 concerns PyTorch v2.8.0 exhibiting unexpected behavior when using torch.rot90 and torch.randn_like together. The primary sources state the issue exists in PyTorch 2.8.0 and provide no explicit remediation in the initial description. Some connected advisories note that patched versi...