vLLM Authentication Bypass via ASGI Trust (0.3.0-0.22.0)
CVE-2026-48746 Published on June 22, 2026

vLLM: OpenAI auth bypass
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.

NVD

Vulnerability Analysis

CVE-2026-48746 is exploitable with network access, and does not require authorization privileges or user interaction. This vulnerability is considered to have a low attack complexity. An automatable proof of concept (POC) exploit exists. The potential impact of an exploit of this vulnerability is considered to have a high impact on confidentiality, with no impact on integrity, and a high impact on availability.

Attack Vector:
NETWORK
Attack Complexity:
LOW
Privileges Required:
NONE
User Interaction:
NONE
Scope:
UNCHANGED
Confidentiality Impact:
HIGH
Integrity Impact:
NONE
Availability Impact:
HIGH

Weakness Type

What is a HTTP Request Smuggling Vulnerability?

When malformed or abnormal HTTP requests are interpreted by one or more entities in the data flow between the user and the web server, such as a proxy or firewall, they can be interpreted inconsistently, allowing the attacker to "smuggle" a request to one device without the other device being aware of it.

CVE-2026-48746 has been classified to as a HTTP Request Smuggling vulnerability or weakness.


Affected Versions

vllm-project vllm Version >= 0.3.0, < 0.22.0 is affected by CVE-2026-48746