---
id: "GHSA-hfwf-hgvp-4rr8"
kind: "vulnerability"
aliases: ["GHSA-hfwf-hgvp-4rr8"]
title: "Duplicate Advisory: vLLM vulnerable to remote denial of service through forged multimodal UUID cache desynchronization"
severity: "high"
cvss: 7.5
epss: null
exploited: false
status: "withdrawn"
category: "ai-tooling"
products: [{"name": "vllm", "fixed": [], "affected": "< 0.31.0", "ecosystem": "pypi"}]
ecosystems: ["pypi"]
tags: ["ai-tooling", "cwe-345", "llm-serving", "python", "self-hosted", "vllm"]
not_affected_if: ["managed-hosting", "vllm-0.31.0-or-later"]
published_at: "2026-10-21T03:32:14+00:00"
first_seen_at: "2026-10-23T20:45:22+00:00"
updated_at: "2026-10-23T20:55:45+00:00"
sources: [{"source": "ghsa", "url": "https://github.com/advisories/GHSA-hfwf-hgvp-4rr8", "first_seen_at": "2026-10-23T20:45:22+00:00"}]
---

# Duplicate Advisory: vLLM vulnerable to remote denial of service through forged multimodal UUID cache desynchronization

This advisory is withdrawn as a duplicate of GHSA-p92p-rxj5-7p2x; the original text says vLLM before 0.31.0 trusts a client-supplied multimodal uuid as cache identity. An unauthenticated attacker can desynchronize the processor caches and crash the whole serving instance until restart.

_Summary, check and fix written by a model from the sources; labelled as such._

## Affected

- pypi:vllm affected < 0.31.0 fixed ?

## How to check

pip show vllm (affected if the version is below 0.31.0)
Check the installed version against the advisory GHSA-p92p-rxj5-7p2x, which carries the live record

## Fix

Upgrade vllm to 0.31.0 or later, as the original advisory states; confirm the fixed version in GHSA-p92p-rxj5-7p2x

## Timeline

- 2026-10-23T20:45:22+00:00 ghsa https://github.com/advisories/GHSA-hfwf-hgvp-4rr8

## From the sources

## Duplicate Advisory

This advisory has been withdrawn because it is a duplicate of GHSA-p92p-rxj5-7p2x. This link is maintained to preserve external references.

## Original Description
vLLM before 0.31.0 uses a client-supplied multimodal uuid verbatim as cache identity with no consistency check on cache hits, allowing engine crashes. Unauthenticated attackers can desynchronize the P0/P1 processor caches to trigger a fatal token-placeholder mismatch, killing the entire serving instance until restarted.

---

_Free to read at a human pace. An API with keys and an MCP server are in preparation; early access: hello@infrastack.ai_