CVE-2025-46153
Linuxfoundation Pytorch 2.6.0 – 2.7.0
Raw vector
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:NCVSS and EPSS are reproduced from their sources (NVD, FIRST EPSS). Risk Priority is our own derived reading, not an NVD score.
Summary
CVE-2025-46153 is a medium-severity Inefficient CPU Computation (CWE-1176) vulnerability in Linuxfoundation Pytorch. Its CVSS base score is 5.3 (Medium).
Operationally, ranked at the 35th percentile by exploit likelihood (below the median); it is not currently listed in the CISA KEV catalog.
This vulnerability is AI-related — categorised as Deep Learning Frameworks; in the Other ATLAS/OWASP Terms risk domain.
EU & UK References
- 🇪🇺 ENISA EUVD: EUVD-2025-31108
Vulnerability Data
PyTorch 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=True.
- CWE(s)
AI Security AnalysisAI
- AI Category
- Deep Learning Frameworks
- Risk Domain
- Other ATLAS/OWASP Terms
- OWASP Top 10 for LLMs 2025
- None mapped
- Classification Reason
- Matched keywords: pytorch
Related Threats
CVEs Like This One
Affected Assets
Mitigating Controls
Mitigating Controls (NIST CSF 2.0) AI
Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→CSF cross-walk (authority under review) — links open the control.
Monitoring compute resources can detect high CPU usage from inefficient algorithms but does not prevent their use.
Capacity management for compute resources can compensate for or flag inefficient CPU usage but does not address root algorithmic choices.
Mitigating Controls (ISO/IEC 27001:2022 Annex A) AI
Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→ISO cross-walk (authority under review) — links open the control.
Secure development lifecycle includes performance and efficiency reviews that can catch inefficient CPU algorithms.
Secure architecture principles encourage selection of efficient algorithms and resource-conscious design.
Secure coding standards can mandate use of efficient algorithms and prohibit known inefficient patterns.
Capacity management can drive optimization of CPU-intensive algorithms to prevent resource exhaustion.