Silent output corruption via inconsistent Inductor compilation
Published Sep 25, 2025 · Updated Sep 25, 2025
Incorrect calculation in PyTorch 2.8.0 and earlier allows remote attackers to corrupt model output via an Inductor-compiled model. The Inductor-compiled path generates random tensor elements in a different order than eager execution when torch.randn_like follows torch.rot90. Reachability requires an application to compile attacker-supplied model logic containing both operations; the demonstrated consequence is silent output corruption rather than a crash.
Summary
What happened
Incorrect calculation in PyTorch 2.8.0 and earlier allows remote attackers to corrupt model output via an Inductor-compiled model. The Inductor-compiled path generates random tensor elements in a different order than eager execution when torch.randn_like follows torch.rot90. Reachability requires an application to compile attacker-supplied model logic containing both operations; the demonstrated consequence is silent output corruption rather than a crash.
The record
- CVE
- CVE-2025-55552
- Published
- Sep 25, 2025
- Updated
- Sep 25, 2025
- Vendor
- Unknown vendor
- Product
- Unknown product
- Classifications
- CWE-682, CWE-190, T1565
- Attack vector
- network
- Privileges
- unauthenticated
Timeline
How it unfolded
- Sep 25, 2025CVE publishedPublication date reported by the CVE source.
- Sep 25, 2025Record updatedLatest update available in the CVE record.
Exploitability
Present is not the same as exploitable
Compare your product and version with the public record. A matching version still requires validation against your environment.
Is a vulnerable build present?
What conditions does exploitation require?
What is affected?
Affected products and versions are unavailable in this record.
Attacks
What attackers are doing with it
Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.
Weakness, pattern, technique
Public exploit references
- PyTorch issue 147847 reproducerproof of concept · demonstrated
Labels summarize the accepted research assessment. They do not indicate a test against your environment.
Technologies
Your stack
See the directory against your own environment.
Your stack
Check the software in your environment
Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.
Book a demo