CVE-2026-34756 — vLLM Affected by Unauthenticated OOM Denial of Service via Unbounded `n` Parameter in OpenAI API Server

Description (CNA)

vLLM is an inference and serving engine for large language models (LLMs). From 0.1.0 to before 0.19.0, a Denial of Service vulnerability exists in the vLLM OpenAI-compatible API server. Due to the lack of an upper bound validation on the n parameter in the ChatCompletionRequest and CompletionRequest Pydantic models, an unauthenticated attacker can send a single HTTP request with an astronomically large n value. This completely blocks the Python asyncio event loop and causes immediate Out-Of-Memory crashes by allocating millions of request object copies in the heap before the request even reaches the scheduling queue. This vulnerability is fixed in 0.19.0.

Affected (CNA)

Vendor / product	Versions
vllm-project / vllm	affected >= 0.1.0, < 0.19.0

CWE

CWE-770

References

ADP enrichment

CISA-ADP

No English description supplied.

No references supplied.

redhat-SADP

A flaw was found in vLLM, an inference and serving engine for large language models (LLMs). An unauthenticated attacker can exploit this vulnerability by sending a specially crafted HTTP request with an excessively large 'n' parameter to the vLLM OpenAI-compatible API server. This can lead to a Denial of Service (DoS) by consuming excessive memory and blocking the system's event loop, causing the server to crash.