High severity · published May 1, 2025
CVE-2025-23254
NVIDIA TensorRT-LLM for any platform contains a vulnerability in python executor where an attacker may cause a data validation issue by local access to the TRTLLM server
8.8
CVSS
0.3%
EPSS
No
On KEV
Description
NVIDIA TensorRT-LLM for any platform contains a vulnerability in python executor where an attacker may cause a data validation issue by local access to the TRTLLM server. A successful exploit of this vulnerability may lead to code execution, information disclosure and data tampering.
CVSS vector
CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H
Attack vector
Local
Complexity
Low
Privileges
Low
User interaction
None
Confidentiality
High
Integrity
High
Availability
High
Scope
Changed
Exploitation
Is it being used?
Exploit prediction (EPSS)0.3% · higher than 20% of CVEs
Exploited in the wildNo evidence yet
CISA KEVNot listed
Ransomware campaign useNot known
Affected products
Which versions
TE
TensorRT-LLM · NVIDIA
All versions prior to 0.18.2 · source: vendor
References
NVD
https://nvd.nist.gov/vuln/detail/CVE-2025-23254
CVE.org
https://www.cve.org/CVERecord?id=CVE-2025-23254
nvidia.custhelp.com
https://nvidia.custhelp.com/app/answers/detail/a_id/5648
More from TensorRT-LLM
9.8
CVE-2026-24163
NVIDIA TRT-LLM for any platform contains a vulnerability in RPC testing, where an attacker could cause an unsafe deserialization
9.8
CVE-2026-24142
NVIDIA TRT-LLM for any platform contains a deserialization vulnerability and unsafe serialized handle
9.8
CVE-2025-33255
NVIDIA TRT-LLM for any platform contains a vulnerability in MPI server, where an attacker could cause an unsafe deserialization
8.4
CVE-2026-65119
NVIDIA TensorRT-LLM for any platform contains a vulnerability in KvCacheAwareRouter tokenizer loading, where an attacker could cause inclusion of functionality from an untrusted…
8.4
CVE-2026-24233
NVIDIA TensorRT-LLM for Linux contains a vulnerability in the restricted unpickler used for model weight deserialization, where a local, unauthenticated attacker could cause…