Instituto Nacional de ciberseguridad. Sección Incibe
Instituto Nacional de Ciberseguridad. Sección INCIBE-CERT

Go to Calendar     Go to Press Room     Go to Newsletters subscription

Search

CVE-2026-48746

Severity CVSS v4.0:
Pending analysis
Type:
Unavailable / Other
Publication date:
22/06/2026
Last modified:
22/07/2026

Description

vLLM is an inference and serving engine for large language models (LLMs). From 0.3.0 until 0.22.0, a vulnerability in ASGI web servers and starlette's trust on those web servers enables an authentication bypass of the OpenAI API AuthenticationMiddleware. It allows to use the API without providing the configured VLLM_API_KEY or --api-key. This vulnerability is fixed in 0.22.0.

Impact

Vector 3.x
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:H CVSS v3.1 Severity and Metrics:

Base Score: 9.10 CRITICAL
Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:H

Attack Vector (AV): Network
Attack Complexity (AC): Low
Privileges Required (PR): None
User Interaction (UI): None
Scope (S): Unchanged
Confidentiality (C): High
Integrity (I): None
Availability (A): High

Base Score 3.x
9.10
Severity 3.x
CRITICAL

Vulnerable products and versions

CPE From Up to
cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:* 0.3.0 (including) 0.22.0 (excluding)

To consult the complete list of CPE names with products and versions, see this page


References to Advisories, Solutions, and Tools