NVIDIA Triton Inference Server Double Free DoS via Cancelled Stream
CVE-2025-23322 Published on August 6, 2025
NVIDIA Triton Inference Server for Windows and Linux contains a vulnerability where multiple requests could cause a double free when a stream is cancelled before it is processed. A successful exploit of this vulnerability might lead to denial of service.
Vulnerability Analysis
CVE-2025-23322 is exploitable with network access, and does not require authorization privileges or user interaction. This vulnerability is considered to have a low attack complexity. The potential impact of an exploit of this vulnerability is considered to have no impact on confidentiality and integrity, and a high impact on availability.
Weakness Type
What is a Double-free Vulnerability?
The product calls free() twice on the same memory address, potentially leading to modification of unexpected memory locations. When a program calls free() twice with the same argument, the program's memory management data structures become corrupted. This corruption can cause the program to crash or, in some circumstances, cause two later calls to malloc() to return the same pointer. If malloc() returns the same value twice and the program later gives the attacker control over the data that is written into this doubly-allocated memory, the program becomes vulnerable to a buffer overflow attack.
CVE-2025-23322 has been classified to as a Double-free vulnerability or weakness.
Products Associated with CVE-2025-23322
You can be notified by email with stack.watch whenever vulnerabilities like CVE-2025-23322 are published in NVIDIA Triton Inference Server:
Affected Versions
NVIDIA Triton Inference Server Version All versions prior to 25.06 is affected by CVE-2025-23322Exploit Probability
EPSS (Exploit Prediction Scoring System) scores estimate the probability that a vulnerability will be exploited in the wild within the next 30 days. The percentile shows you how this score compares to all other vulnerabilities.