Information disclosure in NVIDIA Triton Inference Server
CVE-2025-23320
NVIDIA Triton Inference Server for Windows and Linux contains a vulnerability in the Python backend, where an attacker could cause the shared memory limit to be exceeded by sending a very large request. A successful exploit of this vulnerability might lead to information disclosure.
Published · last modified .
CVSS v3 metric
CVSS v3 base score 7.5 (High). Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N.
EPSS exploit prediction
EPSS 0.010, 63.6th percentile (scored ); unchanged over 30 days. About EPSS.
Affected products
- NVIDIA Triton Inference Server — versions All versions prior to 25.07
- NVIDIA Triton Inference Server
Weakness classification (CWE)
Public exploits and proof-of-concept code
1 public repository with exploit or proof-of-concept code for this CVE:
- There-was-a-bird/triton-cve-2025-23320 (last pushed 2025-10-28)
Also mentioned in 3 other repositories (ARPSyndicate/cve-scores, barbaraeivyu/Triton-Inference-Server-PoC, mrk336/Hybrid-Cloud-Networking-Considerations). These reference the CVE in a list, notes or a scanner and are not exploit code for it.
References
- nvd.nist.gov/vuln/detail/CVE-2025-23320 (US Government Resource)
- www.cve.org/CVERecord (Third Party Advisory)
- nvidia.custhelp.com/app/answers/detail/a_id/5687 (Vendor Advisory)
This CVE has public exploit evidence. See the latest CVEs with public exploits.