CVE-2025-48943: vLLM allows clients to crash the openai server with invalid regex
Impact
A denial of service bug caused the vLLM server to crash if an invalid regex was provided while using structured output. This vulnerability is similar to GHSA-6qc9-v4r8-22xg, but for regex instead of a JSON schema.
Issue with more details: https://github.com/vllm-project/vllm/issues/17313
Patches
https://github.com/vllm-project/vllm/pull/17623
Other sources
vLLM is an inference and serving engine for large language models (LLMs). Version 0.8.0 up to but excluding 0.9.0 have a Denial of Service (ReDoS) that causes the vLLM server to crash if an invalid regex was provided while using structured output. This vulnerability is similar to GHSA-6qc9-v4r8-22xg/CVE-2025-48942, but for regex instead of a JSON schema. Version 0.9.0 fixes the issue.
— MITRE
Affected Software
Remediation
Recommended actions to resolve this vulnerability, in priority order.
- Upgrade
Upgrade
pip/vllmto a version that resolves this vulnerability.Fixed in 0.9.0 - Upgrade
Upgrade
vllmto a version that resolves this vulnerability.Fixed in 0.9.0
Event History
Frequently Asked Questions
What is the severity of CVE-2025-48943?
CVE-2025-48943 is categorized as a denial of service vulnerability due to an invalid regex input causing the vLLM server to crash.
How do I fix CVE-2025-48943?
To fix CVE-2025-48943, upgrade the vllm package to version 0.9.0 or later.
Which versions of vllm are affected by CVE-2025-48943?
CVE-2025-48943 affects vllm versions between 0.8.0 and 0.9.0, exclusive of 0.9.0.
What is the impact of exploiting CVE-2025-48943?
Exploiting CVE-2025-48943 can lead to a denial of service condition where the vLLM server becomes unresponsive.
Is there a workaround for CVE-2025-48943?
Currently, the best workaround for CVE-2025-48943 is to avoid using invalid regex patterns in structured output.