s2
--:--:--UTC

Searching APEX

Starting…

  1. Searching Threats, IOCs & Threat Intelligence locally
  2. Querying external providers
  3. Asking AI Forensic Validator
  4. Creating new entry from validated hit

0s elapsed

CVE-2026-44222

📛 CVE Title

vLLM: Remote DoS via Special-Token Placeholders

Description

vLLM is an inference and serving engine for large language models (LLMs). From 0.6.1 to before 0.20.0, there is a a Token Injection vulnerability in vLLM’s multimodal processing. Unauthenticated, text-only prompts that spell special tokens are interpreted as control. Image and video placeholder sequences supplied without matching data cause vLLM to index into empty grids during input-position computation, raising an unhandled IndexError and terminating the worker or degrading availability. Multimodal paths that rely on image_grid_thw/video_grid_thw are affected. This vulnerability is fixed in 0.20.0.

Overview

State
PUBLISHED
Assigner (CNA)
GitHub_M
CVSS severity
MEDIUM
CVSS score
CVSS 6.5 / 10 6.5 6.5 / 10
CVSS vector
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H
Effective score
6.5 / 10 MEDIUM source: CNA overview
CWE(s)
CWE-129
Reserved
2026-05-05
Published
2026-05-12 21:57 UTC
Last updated
2026-05-13 14:24 UTC
Source
https://raw.githubusercontent.com/CVEProject/cvelistV5/main/cves/2026/44xxx/CVE-2026-44222.json
Linked Threat
CVE-2026-44222 — CVE-2026-44222

NVD triage scoring NVD CVE 2.0

Layer NVD adds on top of the CNA's CVE record — published / last-modified timestamps, exploitability / impact subscores, and the FIRST.org EPSS probability that this CVE will be exploited in the wild in the next 30 days.

NVD published
2026-05-12 20:16:43 UTC
NVD last modified
2026-05-14 15:38:19 UTC
NVD CVSS v3.1
CVSS 6.5 / 10 6.5 6.5 / 10 MEDIUM source: security-advisories@github.com
NVD CVSS vector
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H
Exploitability subscore
2.8 / 10
Impact subscore
3.6 / 10
EPSS score
0.0001 (probability of exploitation in next 30 days)
EPSS percentile
3.07% vs all CVEs — higher = more likely to be exploited, as of 2026-05-24

NVD / KEV / EPSS data refreshed 2026-05-25 08:42 UTC. Re-run the 🛰 Backfill from NVD button above to refresh.

European Union Vulnerability Database ENISA EUVD

ENISA's official EU repository for curated vulnerability intelligence. Carries a separate identifier (EUVD-YYYY-NNNN) and frequently exposes an earlier-published description + CVSS than NVD does.

EUVD ID
EUVD-2026-29799
Assigner
GitHub_M
Published
May 12, 2026, 7:57:25 PM
Updated
May 13, 2026, 12:24:53 PM
EUVD base score (CVSS 3.1)
6.5 / 10
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H
EUVD-reported EPSS
0.0100
Vendors
vllm-project
Products
vllm (0.6.1, < 0.20.0)
Aliases
GHSA-hpv8-x276-m59f

ENISA description: vLLM is an inference and serving engine for large language models (LLMs). From 0.6.1 to before 0.20.0, there is a a Token Injection vulnerability in vLLM’s multimodal processing. Unauthenticated, text-only prompts that spell special tokens are interpreted as control. Image and video placeholder sequences supplied without matching data cause vLLM to index into empty grids during input-position computation, raising an unhandled IndexError and terminating the worker or degrading availability. Multimodal paths that rely on image_grid_thw/video_grid_thw are affected. This vulnerability is fixed in 0.20.0.

EUVD references (2)

Affected products (1)

VendorProductVersionsPlatforms
vllm-project vllm >= 0.6.1, < 0.20.0 (affected)

Affected products — CPE 2.3 (1) NVD

NVD's normalized CPE 2.3 matchers, used by vendor tools (vulnerability scanners, asset managers) for automated detection. Compare with the CNA's free-text "Affected products" section above.

  • cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*

Vendor references (2)

References embedded in the original CVE record by the assigning CNA.

MITRE references (2) cveawg.mitre.org

Pulled from MITRE's CVE Services API by the 🛰 Backfill from MITRE button.

Web references (3)

DuckDuckGo results ranked by threat-intel / vendor advisory domains. Generated by the 🔎 Find references (web) button above — same flow as the Remediations search.

NVD-tagged references (2)

Reference list NVD curates from the CNA record, vendor advisories, and third-party reports. The tag chips below are NVD's analyst-assigned categories.

Remediations (17)

Remediations are stored against the linked Threat row; the list below is deduplicated across both pages.

  • web:access.redhat.com

    Learn about our open source products, services, and company. You are here

    2026-05-23 20:56 UTC
  • web:helpx.adobe.com

    Adobe has released a security update for Adobe Acrobat and Reader for Windows and macOS. This update addresses critical and important vulnerabilities. Successful exploitation could lead to arbitrary code execution and arbitrary file system read. Adobe is not aware of any exploits in the wild for any of the issues addressed in these updates.

    2026-05-23 20:56 UTC
  • web:krebsonsecurity.com

    Microsoft expects that exploitation is more likely. May's Patch Tuesday is a welcome respite from April, which saw Microsoft fix a near-record 167 security flaws.

    2026-05-23 20:56 UTC
  • web:portal.msrc.microsoft.com

    The Security Update Guide provides information on the latest Microsoft security updates, helping users understand and address potential vulnerabilities effectively.

    2026-05-23 20:56 UTC
  • web:support.servicenow.com

    Overview The advisories below document publicly disclosed Common Vulnerabilities and Exposures ( CVEs ) in the Now Platform by ServiceNow. Because ServiceNow uses various methods to communicate vulnerability information, patches, and other fixes, customers should review family, security patch , and hotfix release notes, which are available at https://docs.servicenow.com, for a complete list of ...

    2026-05-23 20:56 UTC
  • web:www.bleepingcomputer.com

    Today is Microsoft's March 2026 Patch Tuesday with security updates for 79 flaws, including 2 publicly disclosed zero-day vulnerabilities.

    2026-05-23 20:56 UTC
  • web:www.crowdstrike.com

    Microsoft's April 2026 Patch Tuesday addresses 164 CVEs , featuring 8 Critical vulnerabilities, one exploited zero-day, and one disclosed zero-day.

    2026-05-23 20:56 UTC
  • web:www.linkedin.com

    Oracle has released its January 2026 Critical Patch Update (CPU), a wide-ranging security bulletin that addresses 337 newly identified vulnerabilities across a broad spectrum of its enterprise ...

    2026-05-23 20:56 UTC
  • web:www.oracle.com

    This Critical Patch Update contains 481 new security patches across the product families listed below. Please note that an MOS note summarizing the content of this Critical Patch Update and other Oracle Software Security Assurance activities is located at April 2026 Critical Patch Update: Executive Summary and Analysis.

    2026-05-23 20:56 UTC
  • web:www.virustotal.com

    VirusTotal is a platform for scanning files and URLs for viruses, malware, and other threats using multiple antivirus engines.

    2026-05-23 20:56 UTC
  • web:blog.qualys.com

    Oracle released its second quarterly edition of this year's Critical Patch Update. The update received patches for 481 security vulnerabilities.

    2026-05-26 02:53 UTC
  • web:community.ui.com

    Published: May 21, 2026 Updated: May 22, 2026 Version: 1.1 Revision: 1.1 Summary 1 of 5 A malicious actor with access to the network and high privileges could exploit an Improper Input Validation vulnerability found in UniFi OS devices to execute a Command Injection. Affected Products: UniFi OS Server (Version 5.0.6 and earlier) Mitigation : Update your UniFi OS Server to Version 5.0.8 or later ...

    2026-05-26 02:53 UTC
  • web:oracle-base.com

    This post describes how to identify the required patches for your Oracle products.

    2026-05-26 02:53 UTC
  • web:www.bugcrowd.com

    Vulnerability mitigation is typically considered a temporary or interim solution. While mitigation measures can reduce the immediate risk associated with vulnerabilities, they may not provide a permanent fix . Organizations should aim to prioritize and plan for complete vulnerability remediation whenever feasible and allocate resources accordingly.

    2026-05-26 02:53 UTC
  • web:www.microsoft.com

    Security Update Guide Notifications Microsoft's free Security Update Guide Notifications provide links to security-related software updates and notification of re-released security updates. These notifications are sent via email throughout the month as needed.

    2026-05-26 02:53 UTC
  • web:www.securityweek.com

    Oracle on Tuesday announced the release of 481 new security patches as part of its April 2026 Critical Patch Update (CPU). Across the 28 product families that received security updates, more than 300 patches address vulnerabilities that are remotely exploitable without authentication.

    2026-05-26 02:53 UTC
  • web:zecurit.com

    Get the complete breakdown of Microsoft's May 2026 Patch Tuesday. We analyze the latest security updates and all critical CVEs .

    2026-05-26 02:53 UTC

AI Forensic Analysis

Only Available for Registered Users. Sign in to view.

Raw JSON

The full cvelistV5 record. Download as CVE-2026-44222.json.

{
  "containers": {
    "adp": [
      {
        "metrics": [
          {
            "other": {
              "content": {
                "id": "CVE-2026-44222",
                "options": [
                  {
                    "Exploitation": "none"
                  },
                  {
                    "Automatable": "no"
                  },
                  {
                    "Technical Impact": "partial"
                  }
                ],
                "role": "CISA Coordinator",
                "timestamp": "2026-05-13T12:24:39.409933Z",
                "version": "2.0.3"
              },
              "type": "ssvc"
            }
          }
        ],
        "providerMetadata": {
          "dateUpdated": "2026-05-13T12:24:53.560Z",
          "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
          "shortName": "CISA-ADP"
        },
        "title": "CISA ADP Vulnrichment"
      }
    ],
    "cna": {
      "affected": [
        {
          "product": "vllm",
          "vendor": "vllm-project",
          "versions": [
            {
              "status": "affected",
              "version": ">= 0.6.1, < 0.20.0"
            }
          ]
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "value": "vLLM is an inference and serving engine for large language models (LLMs). From 0.6.1 to before 0.20.0, there is a a Token Injection vulnerability in vLLM\u2019s multimodal processing. Unauthenticated, text-only prompts that spell special tokens are interpreted as control. Image and video placeholder sequences supplied without matching data cause vLLM to index into empty grids during input-position computation, raising an unhandled IndexError and terminating the worker or degrading availability. Multimodal paths that rely on image_grid_thw/video_grid_thw are affected. This vulnerability is fixed in 0.20.0."
        }
      ],
      "metrics": [
        {
          "cvssV3_1": {
            "attackComplexity": "LOW",
            "attackVector": "NETWORK",
            "availabilityImpact": "HIGH",
            "baseScore": 6.5,
            "baseSeverity": "MEDIUM",
            "confidentialityImpact": "NONE",
            "integrityImpact": "NONE",
            "privilegesRequired": "LOW",
            "scope": "UNCHANGED",
            "userInteraction": "NONE",
            "vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H",
            "version": "3.1"
          }
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-129",
              "description": "CWE-129: Improper Validation of Array Index",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "dateUpdated": "2026-05-12T19:57:25.336Z",
        "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "shortName": "GitHub_M"
      },
      "references": [
        {
          "name": "https://github.com/vllm-project/vllm/security/advisories/GHSA-hpv8-x276-m59f",
          "tags": [
            "x_refsource_CONFIRM"
          ],
          "url": "https://github.com/vllm-project/vllm/security/advisories/GHSA-hpv8-x276-m59f"
        },
        {
          "name": "https://github.com/vllm-project/vllm/issues/32656",
          "tags": [
            "x_refsource_MISC"
          ],
          "url": "https://github.com/vllm-project/vllm/issues/32656"
        }
      ],
      "source": {
        "advisory": "GHSA-hpv8-x276-m59f",
        "discovery": "UNKNOWN"
      },
      "title": "vLLM: Remote DoS via Special-Token Placeholders"
    }
  },
  "cveMetadata": {
    "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
    "assignerShortName": "GitHub_M",
    "cveId": "CVE-2026-44222",
    "datePublished": "2026-05-12T19:57:25.336Z",
    "dateReserved": "2026-05-05T15:42:40.518Z",
    "dateUpdated": "2026-05-13T12:24:53.560Z",
    "state": "PUBLISHED"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.2"
}