PyTorch — CVE history (PyPI)

PyTorch

32 CVEs affect the PyTorch PyPI package (highest CVSS 9.8). Latest disclosed: 2026-03-22. Full CVE history sourced from NVD.

Summary

Package
PyTorch (PyPI)
Total CVEs
32
Actively exploited (CISA KEV)
0
Highest CVSS
9.8
Latest disclosed
2026-03-22

Recent CVEs (top 20)

CVESeverityCVSSKEVPublishedSummary
CVE-2026-4538Medium5.32026-03-22A vulnerability was identified in PyTorch 2.10.0.
CVE-2026-24747High8.82026-01-27PyTorch is a Python package that provides tensor computation.
CVE-2025-63396Low3.32025-11-12An issue was discovered in PyTorch v2.5 and v2.7.1.
CVE-2025-55560High7.52025-09-25An issue in pytorch v2.7.0 can lead to a Denial of Service (DoS) when a PyTorch model consists of torch.Tensor.to_sparse() and torch.Tensor.to_dense() and is compiled by Inductor.
CVE-2025-55558High7.52025-09-25A 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-2025-55557High7.52025-09-25A 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-2025-55554Medium5.32025-09-25pytorch v2.8.0 was discovered to contain an integer overflow in the component torch.nan_to_num-.long().
CVE-2025-55553High7.52025-09-25A syntax error in the component proxy_tensor.py of pytorch v2.7.0 allows attackers to cause a Denial of Service (DoS).
CVE-2025-55552High7.52025-09-25pytorch v2.8.0 was discovered to display unexpected behavior when the components torch.rot90 and torch.randn_like are used together.
CVE-2025-55551High7.52025-09-25An issue in the component torch.linalg.lu of pytorch v2.8.0 allows attackers to cause a Denial of Service (DoS) when performing a slice operation.
CVE-2025-46153Medium5.32025-09-25PyTorch before 3.7.0 has a bernoulli_p 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 fallback_random=T…
CVE-2025-46152Medium5.32025-09-25In PyTorch before 2.7.0, bitwise_right_shift produces incorrect output for certain out-of-bounds values of the "other" argument.
CVE-2025-46150Medium5.32025-09-25In PyTorch before 2.7.0, when torch.compile is used, FractionalMaxPool2d has inconsistent results.
CVE-2025-46149Medium5.32025-09-25In PyTorch before 2.7.0, when inductor is used, nn.Fold has an assertion error.
CVE-2025-46148Medium5.32025-09-25In PyTorch through 2.6.0, when eager is used, nn.PairwiseDistance(p=2) produces incorrect results.
CVE-2025-4287Low3.32025-05-05A vulnerability was found in PyTorch 2.6.0+cu124.
CVE-2025-32434Critical9.82025-04-18PyTorch is a Python package that provides tensor computation with strong GPU acceleration and deep neural networks built on a tape-based autograd system.
CVE-2025-3730Low3.32025-04-16A vulnerability, which was classified as problematic, was found in PyTorch 2.6.0.
CVE-2025-3136Low3.32025-04-03A vulnerability, which was classified as problematic, has been found in PyTorch 2.6.0.
CVE-2025-3121Low3.32025-04-02A vulnerability classified as problematic has been found in PyTorch 2.6.0.

All-time worst (top 10 by CVSS)

CVESeverityCVSSKEVPublishedSummary
CVE-2025-32434Critical9.82025-04-18PyTorch is a Python package that provides tensor computation with strong GPU acceleration and deep neural networks built on a tape-based autograd system.
CVE-2024-48063Critical9.82024-10-29In PyTorch <=2.4.1, the RemoteModule has Deserialization RCE.
CVE-2022-45907Critical9.82022-11-26In PyTorch before trunk/89695, torch.jit.annotations.parse_type_line can cause arbitrary code execution because eval is used unsafely.
CVE-2026-24747High8.82026-01-27PyTorch is a Python package that provides tensor computation.
CVE-2024-31583High7.82024-04-17Pytorch before version v2.2.0 was discovered to contain a use-after-free vulnerability in torch/csrc/jit/mobile/interpreter.cpp.
CVE-2025-55560High7.52025-09-25An issue in pytorch v2.7.0 can lead to a Denial of Service (DoS) when a PyTorch model consists of torch.Tensor.to_sparse() and torch.Tensor.to_dense() and is compiled by Inductor.
CVE-2025-55558High7.52025-09-25A 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-2025-55557High7.52025-09-25A 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-2025-55553High7.52025-09-25A syntax error in the component proxy_tensor.py of pytorch v2.7.0 allows attackers to cause a Denial of Service (DoS).
CVE-2025-55552High7.52025-09-25pytorch v2.8.0 was discovered to display unexpected behavior when the components torch.rot90 and torch.randn_like are used together.