vLLM versions >= 0.6.3 and < 0.9.0 contain multiple regular expression denial of service (ReDoS) vulnerabilities. Several regex patterns — in vllm/lora/utils.py, the phi4mini tool parser, and the OpenAI-compatible serving chat endpoint — are susceptible to catastrophic backtracking. An attacker submitting crafted input with nested or repeated structures can trigger severe CPU consumption and performance degradation, resulting in denial of service.
https://github.com/vllm-project/vllm/security/advisories/GHSA-j828-28rj-hfhp
Published: 2026-06-20
Updated: 2026-06-22
Base Score: 4
Vector: CVSS2#AV:N/AC:L/Au:S/C:N/I:N/A:P
Severity: Medium
Base Score: 4.3
Vector: CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:L
Severity: Medium
Base Score: 5.3
Vector: CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:N/VI:N/VA:L/SC:N/SI:N/SA:N
Severity: Medium