Inference infrastructure

CVE-2025-23331 affects NVIDIA Triton Inference Server for Windows and Linux. A user who sends an invalid request can cause a memory allocation with an excessive size value, leading to a segmentation fault. A successful exploit might lead to denial of service.

NVIDIA Triton Inference Server for Windows and Linux contains an integer overflow vulnerability (CWE-190) that an attacker can trigger through specially crafted inputs. A successful exploit might lead to denial of service and data tampering. The NVD assessment is not yet provided, and no affected versions are listed in the source.

NVIDIA Triton Inference Server for Windows and Linux contains a vulnerability, CVE-2025-23326, where an attacker could cause an integer overflow through a specially crafted input. A successful exploit might lead to denial of service. The NVD assessment is not yet provided.

NVIDIA Triton Inference Server for Windows and Linux contains a vulnerability (CWE-674, Uncontrolled Recursion) where an attacker can trigger uncontrolled recursion through a specially crafted input. A successful exploit might lead to denial of service. NVD has not yet provided an assessment, and the source does not list affected versions.

CVE-2025-23324 affects NVIDIA Triton Inference Server for Windows and Linux. A user who sends an invalid request can cause an integer overflow or wraparound (CWE-190), leading to a segmentation fault. A successful exploit might lead to denial of service.

NVIDIA Triton Inference Server for Windows and Linux contains CVE-2025-23323, an integer overflow or wraparound that a user can trigger by sending an invalid request. The overflow leads to a segmentation fault, and a successful exploit might cause denial of service. The entry is classified as CWE-190 and was published to NVD on 08/06/2025.

NVIDIA Triton Inference Server for Windows and Linux contains a double free vulnerability (CWE-415). Multiple requests could trigger it when a stream is cancelled before it is processed. A successful exploit might lead to denial of service.

NVIDIA Triton Inference Server for Windows and Linux contains a vulnerability, tracked as CVE-2025-23321 and classified as CWE-369 (Divide By Zero). A user can trigger a divide by zero by sending an invalid request. A successful exploit might lead to denial of service.

NVIDIA Triton Inference Server for Windows and Linux contains a vulnerability in the Python backend. An attacker could exceed the shared memory limit by sending a very large request, and a successful exploit might lead to information disclosure. The weakness is classified as CWE-209, Generation of Error Message Containing Sensitive Information.

NVIDIA Triton Inference Server for Windows and Linux contains an out-of-bounds write in its Python backend, which an attacker can trigger by sending a request. The source says a successful exploit might lead to remote code execution, denial of service, data tampering, or information disclosure. The NVD assessment is not yet provided, and the record is tagged CWE-787 and CWE-805.

NVIDIA Triton Inference Server for Windows and Linux contains a vulnerability in the Python backend, where an attacker could cause an out-of-bounds write (CWE-787). A successful exploit might lead to code execution, denial of service, data tampering, and information disclosure. NVD has not yet provided an assessment.