CVE-2026-100652

Description

A flaw was found in vLLM. A remote attacker can trigger a Denial of Service (DoS) by sending requests containing stop token identifiers that exceed the supported vocabulary limits. Because the application's service frontends fail to properly validate these identifiers before internal processing, the out-of-bounds values cause a fatal indexing failure during graphics processing unit (GPU) operations, crashing the core service and requiring a full restart.

Statement

This vulnerability is rated Important rather than Moderate because a remote, unauthenticated caller can crash the inference service with a single malformed generation request, resulting in a persistent denial of service. In Red Hat Enterprise Linux AI and OpenShift AI environments where vLLM exposes model serving endpoints over HTTP or gRPC, the frontend fails to validate stopping token identifiers against the active model vocabulary. When submitted with a minimum token threshold, these invalid values cause a fatal GPU execution failure that forces the entire inference engine offline until manually restarted, though the flaw cannot be leveraged for data compromise or code execution.

Mitigation

Restrict network access to vLLM HTTP and gRPC service endpoints to trusted networks and authenticated clients only.

In host-based environments, restrict the serving port (typically TCP port 8000) using firewalld:

firewall-cmd --permanent --zone=trusted --add-source=<trusted_ip_or_subnet>
firewall-cmd --permanent --zone=public --remove-port=8000/tcp
firewall-cmd --reload

In Red Hat OpenShift AI deployments, apply a Kubernetes NetworkPolicy to the model-serving namespace to ensure only authorized ingress controllers or trusted application pods can reach the inference containers.

Warning: Reloading firewall rules with `firewall-cmd --reload` applies changes immediately and may interrupt active network connections.

Caveats: Network-level controls limit exposure to external untrusted callers, but they do not protect against crafted requests submitted by authorized users or internal services that consume the inference API.

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).

CVSS v3 Score Breakdown

Red HatNVDcve.org
Base Score7.57.55.9
Attack VectorNetworkNetworkNetwork
Attack ComplexityLowLowHigh
Privileges RequiredNoneNoneNone
User InteractionNoneNoneNone
ScopeUnchangedUnchangedUnchanged
ConfidentialityNoneNoneNone
Integrity ImpactNoneNoneNone
Availability ImpactHighHighHigh

Vector

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

NVD: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H

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

Understanding the Weakness (CWE)

Other

Technical Impact: Varies by Context

Frequently Asked Questions

Want to get errata notifications? Sign up here.