---
id: CVE-2026-105753
title: vLLM is an inference and serving engine for large language models
summary: >-
  vLLM is an inference and serving engine for large language models. Prior to
  0.28.0, the default mirrored multimodal LRU cache can commit a media hash in
  the frontend sender cache during multimodal rendering and before engine
  admission, w…
severity: medium
cvss: 6.5
cvssVector: 'CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H'
cwe:
  - CWE-617
vendor: vllm
product: vllm
affected:
  - vllm < 0.28.0
patched:
  - vllm 0.28.0
published: '2026-10-05'
updated: '2026-10-08'
sourceUpdated: '2026-10-08T01:29:33.007'
source: NVD
sourceUrl: 'https://nvd.nist.gov/vuln/detail/CVE-2026-105753'
references:
  - url: >-
      https://github.com/vllm-project/vllm/commit/396204230423b7cc6798300926b8fa30190d26a9
    label: security-advisories@github.com
  - url: 'https://github.com/vllm-project/vllm/pull/46747'
    label: security-advisories@github.com
  - url: 'https://github.com/vllm-project/vllm/pull/51897'
    label: security-advisories@github.com
  - url: 'https://github.com/vllm-project/vllm/releases/tag/v0.28.0'
    label: security-advisories@github.com
  - url: >-
      https://github.com/vllm-project/vllm/security/advisories/GHSA-ph3r-5jfg-f84f
    label: security-advisories@github.com
  - url: 'https://github.com/advisories/GHSA-ph3r-5jfg-f84f'
  - url: 'https://github.com/vllm-project/vllm'
  - url: >-
      https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-105753.json
  - url: 'https://access.redhat.com/security/cve/CVE-2026-105753'
  - url: 'https://bugzilla.redhat.com/show_bug.cgi?id=2546180'
  - url: 'https://www.cve.org/CVERecord?id=CVE-2026-105753'
  - url: 'https://nvd.nist.gov/vuln/detail/CVE-2026-105753'
tags:
  - nvd
  - cve.org
  - ghsa
  - pip
  - osv
  - csaf
  - vex
  - red-hat
epss: 0.00427
epssPercentile: 0.36807
aliases:
  - GHSA-ph3r-5jfg-f84f
  - PYSEC-2026-4197
ecosystem: pip
ssvc:
  exploitation: none
  automatable: 'no'
  technicalImpact: partial
  timestamp: '2026-10-06T13:21:53.802331Z'
ingestedAt: '2026-10-05T23:36:21.155Z'
---

## Overview

vLLM is an inference and serving engine for large language models. Prior to 0.28.0, the default mirrored multimodal LRU cache can commit a media hash in the frontend sender cache during multimodal rendering and before engine admission, while the engine receiver cache never receives the payload if that request is rejected. A later request reusing the same media hash causes MultiModalProcessorSenderCache to send no payload and MultiModalReceiverCache to reach an assertion with the message "Expected a cached item," producing a shared-service availability failure. This issue is fixed in version 0.28.0.

## Affected

- `vllm < 0.28.0`

## Remediation

Upgrade past the affected range:

- `vllm 0.28.0`

## Package advisory (CVE-2026-105753)

Affected packages:

- `vllm < 0.28.0`

Patched in:

- `vllm 0.28.0`

Source: https://github.com/advisories/GHSA-ph3r-5jfg-f84f

## Vendor advisories

- **Red Hat VEX** · Moderate · affected: Red Hat AI Inference Server, Red Hat Enterprise Linux AI (RHEL AI) 3, Red Hat OpenShift AI (RHOAI) · no fix planned: Red Hat AI Inference Server, Red Hat Enterprise Linux AI (RHEL AI) 3, Red Hat OpenShift AI (RHOAI) · updated 2026-10-06 · [vex](https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-105753.json)