CVE-2025-48887
Last modified
CVE-2025-48887 is a medium-severity vulnerability rated 6.5/10 on the CVSS scale. vLLM, an inference and serving engine for large language models (LLMs), has a Regular Expression Denial of Service (ReDoS) vulnerability in the file `vllm/entrypoints/openai/tool_parsers/pythonic_tool_parser.py` of versions 0.6.4 up to but excluding 0.9.0. The root cause is the use of a highly complex and nested regular expression for tool call detection, which can be exploited by an attacker to cause severe performance degradation or make the service unavailable. EPSS estimates a 0.43% chance of exploitation in the next 30 days.
Description
vLLM, an inference and serving engine for large language models (LLMs), has a Regular Expression Denial of Service (ReDoS) vulnerability in the file `vllm/entrypoints/openai/tool_parsers/pythonic_tool_parser.py` of versions 0.6.4 up to but excluding 0.9.0. The root cause is the use of a highly complex and nested regular expression for tool call detection, which can be exploited by an attacker to cause severe performance degradation or make the service unavailable. The pattern contains multiple nested quantifiers, optional groups, and inner repetitions which make it vulnerable to catastrophic backtracking. Version 0.9.0 contains a patch for the issue.
Metrics
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H
Weakness Enumeration
Affected Software
| Vendor | Product | Versions |
|---|---|---|
| Vllm | Vllm | >= 0.6.4, < 0.9.0 |
References
- https://github.com/vllm-project/vllm/pull/18454Issue Tracking, Patch
- https://github.com/vllm-project/vllm/security/advisories/GHSA-w6q7-j642-7c25Exploit, Vendor Advisory
- https://github.com/vllm-project/vllm/security/advisories/GHSA-w6q7-j642-7c25Exploit, Vendor Advisory
Timeline
- Published
- Last Modified
- Status
- Analyzed
Frequently Asked Questions
What is CVE-2025-48887?
How severe is CVE-2025-48887?
How do I fix CVE-2025-48887?
How Strix Helps
- How Strix found a critical auth bypass in etcdStrix autonomously discovered a critical authentication bypass in etcd, later designated CVE-2026-33413.
- Autonomous PentestingAI agents that find and validate exploitable vulnerabilities like this one across your applications.
- PR ReviewsPentest every pull request so vulnerable code is caught before it ships to production.
- AI Penetration TestingHow AI-driven penetration testing continuously covers your attack surface.
Related CVEs from 2025
- CVE-2025-48881Valtimo is a platform for Business Process Automation. In ve…8.3
- CVE-2025-48882PHPOffice Math is a library that provides a set of classes t…8.7
- CVE-2025-48883Chrome PHP allows users to start playing with chrome/chromiu…5.3
- CVE-2025-48884Galette is a membership management web application for non p…6.1
- CVE-2025-48885application-urlshortener create shortened URLs for XWiki pag…5.7
- CVE-2025-48886Hydra is a layer-two scalability solution for Cardano. Prior…4.8
- CVE-2025-48888Deno is a JavaScript, TypeScript, and WebAssembly runtime. S…5.3
- CVE-2025-48889Gradio is an open-source Python package that allows quick bu…7.5
- CVE-2025-4889A vulnerability has been found in code-projects Tourism Mana…7.8
- CVE-2025-48890WRH-733GBK and WRH-733GWH contain an improper neutralization…9.8
- CVE-2025-48891A vulnerability exists in Advantech iView that could allow f…7.6
- CVE-2025-4890A vulnerability was found in code-projects Tourism Managemen…7.8
Are you affected by CVE-2025-48887?
Run a free Strix scan to check your systems for this vulnerability.
Scan your code nowSource: NVD / NIST