CVE-2026-56340: Vllm VLLM Introduced Enhanced Protection For Vulnerability

HIGH
Advisory Identifier: GHSA-mcmc-2m55-j8jj
Share
CVSS 8.8
CWE-123 CWE-502 CWE-787 CWE-20
Published
Advisory Summary

CVE-2026-56340 is a HIGH (CVSS 8.8) vllm introduced enhanced protection for vulnerability (CWE-123: Write-what-where Condition) affecting Vllm libraries in the PyPI ecosystem. The issue involves vllm introduced enhanced protection for cve-2025-62164. Remediation is available by upgrading affected packages to version 0.13.0.

CVE-2026-56340 at a Glance

CVE Identifier CVE-2026-56340
GHSA Identifier GHSA-mcmc-2m55-j8jj
Severity HIGH
CVSS v3.1 Base Score 8.8
CWE Classification CWE-123: Write-what-where Condition
Affected Product Vllm
Package Ecosystem PyPI
Published Date
Last Updated

Security Impact

CVSS: 8.8

The security impact metrics below are parsed directly from the CVSS v3.1 vector string (CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H).

Attack Vector Network (AV:N)
Attack Complexity Low (AC:L)
Privileges Required Low (PR:L)
User Interaction None (UI:N)
Scope Unchanged (S:U)
Confidentiality High (C:H)
Integrity High (I:H)
Availability High (A:H)

How to Fix CVE-2026-56340

To remediate this vulnerability, upgrade all affected Vllm dependencies to a patched release. Both direct and transitive dependencies should be checked, as frameworks or security libraries may pull in vulnerable versions transitively.

vllm fixed: 0.13.0 
pip install vllm>=0.13.0
Verify Direct and Transitive Dependencies

Run the following command in your project directory to verify that all resolved dependencies use the patched release: 

pip show vllm

How to Check if You Are Affected

To determine whether your Python environment is affected by CVE-2026-56340, inspect your installed packages using pip:

pip show vllm

Analyze the terminal output to locate any version of Vllm in your dependency tree. If any resolved version matches the vulnerable ranges listed in the table below (for example, versions earlier than 0.13.0), your application is affected.

Note: The pip show vllm command checks currently installed packages in your active Python environment.

Technical Details

Vulnerability Mechanism & Technical Analysis

This vulnerability is classified under CWE-123: Write-what-where Condition, where any condition where the attacker has the ability to write an arbitrary value to an arbitrary location, often as the result of a buffer overflow. In Vllm, software state validation or input boundaries are circumvented during processing. Review the official source advisory statement below for specific affected components, classes, and reproduction details.

Source Advisory Statement

### Summary The fix [here](https://github.com/vllm-project/vllm/pull/27204) for CVE-2025-62164 is not sufficient. The fix only disables prompt embeds by default rather than addressing the root cause, so the DoS vulnerability remains when the feature is enabled. ### Details vLLM's pending change attempts to fix the root cause, which is the missing sparse tensor validation. PyTorch (~v2.0) disables sparse tensor validation (specifically, sparse tensor invariants checks) by default for performance reasons. vLLM is adding the sparse tensor validation to ensure indices are valid, non-negative, and within bounds. These checks help catch malformed tensors. ### PoC NA ### Impact Current fix only added a flag to disable/enable prompt embeds, so by default, prompt embeds feature is disabled in vLLM, which stops DoS attacks through the embeddings. However, It doesn’t address the problem when the flag is enabled and there is still potential for DoS attacks. ### Changes * https://github.com/vllm-project/vllm/pull/30649

CVE-2026-56340 Affected Vllm Versions

1 component(s)
Fixed Versions:
0.13.0
Product / Package Ecosystem Affected Versions Fixed Version
vllm
pypi/vllm
PyPI
≥ 0.10.2
< 0.13.0
0.13.0

References

GitHub Advisory / Wiki ADVISORY
https://github.com/vllm-project/vllm/security/advisories/GHSA-mcmc-2m55-j8jj
Vllm Source Repository REPOSITORY
https://github.com/vllm-project/vllm/pull/30649
Vllm Security Information SECURITY
https://access.redhat.com/security/cve/CVE-2026-56340
bugzilla.redhat.com/show_bug.cgi WEB
https://bugzilla.redhat.com/show_bug.cgi?id=2491060
Vllm Source Repository REPOSITORY
https://github.com/pypa/advisory-database/tree/main/vulns/vllm/PYSEC-2026-250.yaml
Vllm Source Repository REPOSITORY
https://github.com/vllm-project/vllm
Vllm Security Information SECURITY
https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-56340.json
www.vulncheck.com/advisories/vllm-den... WEB
https://www.vulncheck.com/advisories/vllm-denial-of-service-via-unvalidated-multimodal-embeddings

Frequently Asked Questions

What is CVE-2026-56340?
CVE-2026-56340 is a security vulnerability in Vllm involving vLLM introduced enhanced protection for CVE-2025-62164 under CWE-123: Write-what-where Condition.
Which Vllm versions are affected?
Affected versions include releases prior to 0.13.0 across affected packages: vllm.
What version fixes CVE-2026-56340?
Upgrading to 0.13.0 (or the latest patched release for your respective branch) resolves the vulnerability.
Is vllm affected?
Yes, vllm is affected in releases prior to 0.13.0. Upgrade your dependency to 0.13.0 or later.
What is the CVSS score of CVE-2026-56340?
CVE-2026-56340 has a CVSS v3.1 base score of 8.8 (HIGH) with vector string CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H.
What CWE is associated with CVE-2026-56340?
CVE-2026-56340 is classified under CWE-123: Write-what-where Condition (Any condition where the attacker has the ability to write an arbitrary value to an arbitrary location, often as the result of a buffer overflow.).
How can I check PyPI dependencies for CVE-2026-56340?
Run the dependency tree command: 'pip show vllm'. This command lists direct and transitive dependencies in your build configuration so you can verify if a vulnerable version is resolved.
Machine-readable OSV JSON OSV v1.5 Schema
Normalized schema output
View Raw JSON 
{
  "schema_version" : "1.4.0",
  "id" : "GHSA-mcmc-2m55-j8jj",
  "modified" : "2026-09-11T19:19:10Z",
  "published" : "2026-01-08T21:47:43Z",
  "aliases" : [ "CVE-2026-56340" ],
  "summary" : "vLLM introduced enhanced protection for CVE-2025-62164",
  "details" : "### Summary\nThe fix [here](https://github.com/vllm-project/vllm/pull/27204) for CVE-2025-62164 is not sufficient. The fix only disables prompt embeds by default rather than addressing the root cause, so the DoS vulnerability remains when the feature is enabled.\n\n### Details\nvLLM's pending change attempts to fix the root cause, which is the missing sparse tensor validation.  PyTorch (~v2.0) disables sparse tensor validation (specifically, sparse tensor invariants checks) by default for performance reasons.  vLLM is adding the sparse tensor validation to ensure indices are valid, non-negative, and within bounds.  These checks help catch malformed tensors.\n\n### PoC\nNA\n\n### Impact\nCurrent fix only added a flag to disable/enable prompt embeds, so by default, prompt embeds feature is disabled in vLLM, which stops DoS attacks through the embeddings.  However, It doesn’t address the problem when the flag is enabled and there is still potential for DoS attacks.\n\n### Changes\n\n* https://github.com/vllm-project/vllm/pull/30649",
  "severity" : [ {
    "type" : "CVSS_V3",
    "score" : "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H"
  } ],
  "affected" : [ {
    "package" : {
      "ecosystem" : "PyPI",
      "name" : "vllm"
    },
    "ranges" : [ {
      "type" : "ECOSYSTEM",
      "events" : [ {
        "introduced" : "0.10.2"
      }, {
        "fixed" : "0.13.0"
      } ]
    } ]
  } ],
  "references" : [ {
    "type" : "WEB",
    "url" : "https://github.com/vllm-project/vllm/security/advisories/GHSA-mcmc-2m55-j8jj"
  }, {
    "type" : "WEB",
    "url" : "https://github.com/vllm-project/vllm/pull/30649"
  }, {
    "type" : "WEB",
    "url" : "https://access.redhat.com/security/cve/CVE-2026-56340"
  }, {
    "type" : "WEB",
    "url" : "https://bugzilla.redhat.com/show_bug.cgi?id=2491060"
  }, {
    "type" : "WEB",
    "url" : "https://github.com/pypa/advisory-database/tree/main/vulns/vllm/PYSEC-2026-250.yaml"
  }, {
    "type" : "PACKAGE",
    "url" : "https://github.com/vllm-project/vllm"
  }, {
    "type" : "WEB",
    "url" : "https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-56340.json"
  }, {
    "type" : "WEB",
    "url" : "https://www.vulncheck.com/advisories/vllm-denial-of-service-via-unvalidated-multimodal-embeddings"
  } ],
  "database_specific" : {
    "cwe_ids" : [ "CWE-123", "CWE-20", "CWE-502", "CWE-787" ],
    "severity" : "HIGH",
    "github_reviewed" : true,
    "github_reviewed_at" : "2026-01-08T21:47:43Z",
    "nvd_published_at" : null
  }
}