---
id: CVE-2026-115914
title: vLLM DeepSeek-OCR image processing vulnerable to denial of service
summary: >-
  vLLM is an inference and serving engine for large language models. From 0.11.1
  until 0.32.0, the DeepSeek-OCR, DeepSeek-OCR-2, and Unlimited-OCR image path
  allows a remote client that can reach the OpenAI-compatible API to submit a
  legal…
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'
cvssSource: cna
cwe:
  - CWE-617
vendor: vllm-project
product: vllm
affected:
  - 'vllm >= 0.11.1, < 0.32.0'
ssvc:
  exploitation: none
  automatable: 'no'
  technicalImpact: partial
  timestamp: '2026-10-30T14:23:13.354875Z'
published: '2026-10-26'
updated: '2026-10-30'
sourceUpdated: '2026-10-30T14:23:21.521Z'
source: CVEORG
sourceUrl: 'https://www.cve.org/CVERecord?id=CVE-2026-115914'
references:
  - url: >-
      https://github.com/vllm-project/vllm/security/advisories/GHSA-gq5q-89c7-vhvx
    label: >-
      https://github.com/vllm-project/vllm/security/advisories/GHSA-gq5q-89c7-vhvx
  - url: 'https://github.com/vllm-project/vllm/pull/59417'
    label: 'https://github.com/vllm-project/vllm/pull/59417'
  - url: >-
      https://github.com/vllm-project/vllm/commit/c15672cfb49152ebf65cfd188e71b84593ec5686
    label: >-
      https://github.com/vllm-project/vllm/commit/c15672cfb49152ebf65cfd188e71b84593ec5686
  - url: 'https://github.com/vllm-project/vllm/releases/tag/v0.32.0'
    label: 'https://github.com/vllm-project/vllm/releases/tag/v0.32.0'
  - url: 'https://nvd.nist.gov/vuln/detail/CVE-2026-115914'
  - url: 'https://github.com/vllm-project/vllm'
  - url: 'https://github.com/advisories/GHSA-gq5q-89c7-vhvx'
tags:
  - cve.org
  - osv
  - pip
  - ghsa
epss: 0.00397
epssPercentile: 0.34109
aliases:
  - GHSA-gq5q-89c7-vhvx
ecosystem: pip
patched:
  - vllm 0.32.0
ingestedAt: '2026-10-26T17:33:33.055Z'
---

## Overview

vLLM is an inference and serving engine for large language models. From 0.11.1 until 0.32.0, the DeepSeek-OCR, DeepSeek-OCR-2, and Unlimited-OCR image path allows a remote client that can reach the OpenAI-compatible API to submit a legal image whose normalized pixels sum to zero when that image is alone in its per-step encoder batch. _parse_and_validate_image_input returns no embedding, embed_multimodal propagates that value, and sanity_check_mm_encoder_outputs triggers a fatal assertion that shuts down the engine for all users. This issue is fixed in version 0.32.0.

## Affected

- `vllm >= 0.11.1, < 0.32.0`

## Remediation

Refer to the linked advisories for vendor-supplied fixes and affected version ranges.

## Package advisory (CVE-2026-115914)

Affected packages:

- `vllm >= 0.11.1, < 0.32.0`

Patched in:

- `vllm 0.32.0`

Source: https://osv.dev/vulnerability/GHSA-gq5q-89c7-vhvx