CVE-2026-126396

Description

A flaw was found in vLLM. This flaw allows a remote attacker to cause a Denial of Service (DoS) by submitting requests with excessively large frame dimension values to the MiniMax M3 video processor. Because the processor fails to validate these parameters against maximum bounds, the application excessively enlarges decoded video frames. This behavior leads to disproportionate CPU and memory consumption, potentially crashing worker processes or making the service unresponsive.

Statement

This vulnerability is rated as Moderate impact for Red Hat deployments because, although it can be triggered over the network by unauthenticated API callers, the impact is strictly limited to service degradation or denial of service through resource exhaustion without affecting data confidentiality or integrity. Exposure is constrained to environments actively serving MiniMax M3 vision-language workloads that accept video inputs. Typical inference deployments not serving this specific multimodal architecture or operating behind authenticated API gateways with restricted client access face minimal operational risk.

Mitigation

Restrict network access to the vLLM service port to trusted clients or localhost using firewall rules:

firewall-cmd --permanent --zone=public --remove-port=8000/tcp
firewall-cmd --permanent --zone=trusted --add-source=192.168.1.0/24
firewall-cmd --reload

Warning: Reloading the firewall configuration will momentarily interrupt active network connections. Verify network and management connectivity before applying firewall changes.

Caveats: Restricting network exposure prevents untrusted external clients from reaching the API, but authorized clients within the trusted subnet can still submit requests that trigger resource exhaustion. Workloads that do not serve MiniMax M3 VL models are unaffected.

Common Vulnerability Scoring System (CVSS) Score Details

Info alert:Important note

CVSS scores for open source components depend on vendor-specific factors (e.g. version or build chain). Therefore, Red Hat's score and impact rating can be different from NVD and other vendors. Red Hat remains the authoritative CVE Naming Authority (CNA) source for its products and services (see Red Hat classifications).

The following CVSS metrics and score provided are preliminary and subject to review.

CVSS v3 Score Breakdown

Red HatNVDcve.org
Base Score5.3N/A5.3
Attack VectorNetworkN/ANetwork
Attack ComplexityLowN/ALow
Privileges RequiredNoneN/ANone
User InteractionNoneN/ANone
ScopeUnchangedN/AUnchanged
ConfidentialityNoneN/ANone
Integrity ImpactNoneN/ANone
Availability ImpactLowN/ALow

Vector

Red Hat: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L

cve.org: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L

Understanding the Weakness (CWE)

Availability

Technical Impact: DoS: Resource Consumption (CPU); DoS: Resource Consumption (Memory); DoS: Resource Consumption (Other)

When allocating resources without limits, an attacker could prevent other systems, applications, or processes from accessing the same type of resource. It can be easy for an attacker to consume many resources by rapidly making many requests or causing larger resources to be used than is needed.

Frequently Asked Questions

Want to get errata notifications? Sign up here.