Your RSA-2048 keys break in 2030. Find every one of them before attackers do.
🐍
🐍 PyPI
Not in CISA KEV
HIGH severity

vLLM introduced enhanced protection for CVE-2025-62164GHSA-mcmc-2m55-j8jj

HIGHFix: vllm-project/vllm#30649

GHSA-mcmc-2m55-j8jj is a high-severity (CVSS 8.8) Improper Input Validation vulnerability in vllm. A fix is available for vllm — see the affected versions and patch details below.

Also known asCVE-2026-56340PYSEC-2026-250
Published
Updated
Affected
1 pkg
Patched
1 / 1
Exploits
None indexed
Exploitation data as of Oct 28, 2026 · OSV.dev, NVD, FIRST.org (EPSS)

Exploitation Status

No confirmed exploitation observed yet

  • A successful exploit gives an attacker total control of the affected component, not partial access.
  • CISA’s own triage has not observed active exploitation or public proof-of-concept code for this CVE as of its last assessment.

Exploitation and automatability from CISA’s SSVC triage for GHSA-mcmc-2m55-j8jj.

EPSS Exploitation Probability

via FIRST.org ↗
0.6%probability of exploitation in next 30 days
Lower Risk0.00%
Lower risk than most CVEs51th percentile — riskier than 51% of all scored CVEsHighest risk
0.00%0.38%0.76%1.14%0.4%0.4%0.6%0.6%0.6%Jul 26Sep 26Oct 26

Probability of exploitation in the next 30 days, from FIRST.org EPSS.

How urgent is this, really

GHSA-mcmc-2m55-j8jj by exploitation likelihood (EPSS) against impact (CVSS). Outside the shaded patch-first corner.

Where this sits among everything scored

Of 397,508 CVEs with a current EPSS score, this one falls in the < 10% band (highlighted). Counts from FIRST.org, log-scaled.

Real-World Exposure

1 pkg affected
🐍vllm

Real-time download stats are indexed for npm and PyPI packages. This vulnerability affects PyPI packages — download data is not available via public APIs for these ecosystems.

Description

Summary

The fix here 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

Affected Packages

1 total 1 fixed
EcosystemPackageVulnerable rangeFix
🐍PyPIvllm≥ 0.10.2&&< 0.13.00.13.0pip install --upgrade 'vllm==0.13.0'

Affected Products

1 product · 1 configurations
Application
vllmvllm
≥ 0.10.2 && < 0.13.0
range

Detection & mitigation playbook

Open-source dependency

  1. Detect

    Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for vllm, including transitive dependencies — a direct dependency you never call can still pull in a vulnerable version.

  2. Fix

    Update vllm to 0.13.0 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms GHSA-mcmc-2m55-j8jj is resolved across your whole dependency graph.

  3. Workarounds

    Cap what an attacker can consume: apply request size, rate and timeout limits in front of the affected component, and run it with memory and CPU limits so exhaustion degrades one worker rather than the whole service.

Fixing This On Your OS

If you run this on a Linux distribution, patch through your package manager against the distro's own security advisory below — it tracks the exact backported fix for your release, which can ship on a different timeline (and sometimes a different severity) than the upstream project.

Red HatImportant

Red Hat rates this issue as having Important impact for affected Red Hat AI Inference Server images shipping vLLM 0.10.2 through 0.13.x when prompt-embeds multimodal embedding support is enabled. Versions outside this range, Red Hat OpenShift AI KServe sidecars, and Red Hat Enterprise Linux AI 3.4 bootc images (vLLM…

Workaround published by Red Hat
Disable prompt-embeds if not required. Restrict who can submit multimodal embedding requests. Apply authentication and rate limits on inference APIs. Upgrade to a fixed vLLM build when available from Red Hat.
Source: Red Hat security advisory for GHSA-mcmc-2m55-j8jj (CC BY 4.0)

Frequently Asked Questions

### 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, a
O3 Security · Impact-Aware SCA

Is GHSA-mcmc-2m55-j8jj in your dependencies?

Find it across PyPI, including transitive dependencies.

Scan my dependencies How O3 SCA works
vLLM introduced enhanced protection for CVE-2025-62164