Read the docs
TI-2026-18296 Aliases CVE-2026-22778GHSA-4r2x-xpjr-7cvv

vLLM leaks a heap address when PIL throws an error

Priority HIGH 67CWE-532Published · Updated

Key facts

Attack vector
Network
Privileges required
None
User interaction
None
Attack complexity
Low
Impact type
Information disclosure, Code execution
Confidentiality · integrity · availability
High · High · High
Fix status
Partial Red Hat: 4 fixed, 1 without a fix

Attack profile from CVSS 3.1 · CNA. Fix status from CNA, GitHub and Red Hat. Impact type from CWE-532, CWE-122 and the CVSS impact.

Summary

vLLM is an inference and serving engine for large language models (LLMs). From 0.8.3 to before 0.14.1, when an invalid image is sent to vLLM's multimodal endpoint, PIL throws an error. vLLM returns this error to the client, leaking a heap address. With this leak, we reduce ASLR from 4 billion guesses to ~8 guesses. This vulnerability can be chained a heap overflow with JPEG2000 decoder in OpenCV/FFmpeg to achieve remote code execution. This vulnerability is fixed in 0.14.1.

Why it matters now

Exploitation None reported CISA SSVC · not in CISA KEV
EPSS 0.18287 97th percentileScored 2026-11-07
Fix Partial vllm (vllm-project): 0.14.1vllm (PyPI): 0.14.1Red Hat: 8 advisoriesRed Hat: 4 fixed, 1 without a fix

SSVC decision points

CISA ADP · SSVC 2.0.3

Exploitation
None
Automatable
Yes
Technical impact
Total

The decision (Track, Attend, Act) also depends on your mission and well-being impact; see CISA's SSVC guide.

Affected products

VendorProductAffected versionsFixed inStatusSource
vllm-projectvllm0.8.3 to before 0.14.10.14.1Patch availableCNA
—vllm (PyPI)0.8.3 to before 0.14.10.14.1Patch availableGHSA-4r2x-xpjr-7cvv
  • vllm Vendorvllm-project Affected 0.8.3 to before 0.14.1 Fixed in 0.14.1 Patch available CNA
  • vllm (PyPI) Vendor— Affected 0.8.3 to before 0.14.1 Fixed in 0.14.1 Patch available GHSA-4r2x-xpjr-7cvv

CPE configurations NVD · 1

CPEVersionsRoleVendor
cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*0.8.3 to before 0.14.1VulnerablevLLM
  • cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:* Versions 0.8.3 to before 0.14.1 Vulnerable ·vLLM

Remediation

  • Fix · CNA vllm (vllm-project) Fixed in 0.14.1
  • Fix · GHSA-4r2x-xpjr-7cvv vllm (PyPI) Fixed in 0.14.1 Advisory →

Red Hat 8 advisories

  • RHSA-2026:30089 Released 2026-06-25 · Red Hat rating Critical
    Red Hat AI Inference Server 3.3.5 (CUDA)
    • Red Hat AI Inference Server 3.3 registry.redhat.io/rhaiis/vllm-cuda-rhel9
  • RHSA-2026:30088 Released 2026-06-25 · Red Hat rating Critical
    Red Hat AI Inference Server 3.3.5 (ROCm)
    • Red Hat AI Inference Server 3.3 registry.redhat.io/rhaiis/vllm-rocm-rhel9
  • RHSA-2026:30087 Released 2026-06-25 · Red Hat rating Critical
    Red Hat AI Inference Server 3.3.5 (Spyre)
    • Red Hat AI Inference Server 3.3 registry.redhat.io/rhaiis/vllm-spyre-rhel9
Show 5 more Red Hat advisories
  • RHSA-2026:19712 Released 2026-05-20 · Red Hat rating Critical
    RHOAI 3.3.3 - Red Hat OpenShift AI
    • Red Hat OpenShift AI 3.3 registry.redhat.io/rhoai/odh-built-in-detector-rhel9, registry.redhat.io/rhoai/odh-dashboard-rhel9, registry.redhat.io/rhoai/odh-data-science-pipelines-argo-argoexec-rhel9 +33
  • RHSA-2026:3782 Released 2026-03-04 · Red Hat rating Critical
    RHOAI 2.25.3 - Red Hat OpenShift AI
    • Red Hat OpenShift AI 2.25 registry.redhat.io/rhoai/odh-caikit-tgis-serving-rhel9, registry.redhat.io/rhoai/odh-codeflare-operator-rhel9, registry.redhat.io/rhoai/odh-dashboard-rhel9 +27
  • RHSA-2026:3713 Released 2026-03-04 · Red Hat rating Critical
    RHOAI 3.3 - Red Hat OpenShift AI
    • Red Hat OpenShift AI 3.3 registry.redhat.io/rhoai/odh-built-in-detector-rhel9, registry.redhat.io/rhoai/odh-dashboard-rhel9, registry.redhat.io/rhoai/odh-data-science-pipelines-argo-argoexec-rhel9 +47
  • RHSA-2026:3462 Released 2026-02-27 · Red Hat rating Critical
    Red Hat AI Inference Server 3.2.2 (ROCm)
    • Red Hat AI Inference Server 3.2 registry.redhat.io/rhaiis/vllm-rocm-rhel9
  • RHSA-2026:3461 Released 2026-02-27 · Red Hat rating Critical
    Red Hat AI Inference Server 3.2.2 (CUDA)
    • Red Hat AI Inference Server 3.2 registry.redhat.io/rhaiis/vllm-cuda-rhel9

Vendor assessments

Red Hat

Red Hat rating: Critical. Red Hat's page for this CVE →

ProductStatusPlan or fix
Red Hat Enterprise Linux AI (RHEL AI) 3AffectedNo fix planned: Out of support scope
Red Hat AI Inference Server 3.2Fixed RHSA-2026:3461, RHSA-2026:3462
Red Hat AI Inference Server 3.3Fixed RHSA-2026:30087, RHSA-2026:30088, RHSA-2026:30089
Red Hat OpenShift AI 2.25Fixed RHSA-2026:3782
Red Hat OpenShift AI 3.3Fixed RHSA-2026:19712, RHSA-2026:3713
Not affected: 4 products
  • Red Hat AI Inference Server
  • Red Hat OpenShift AI (RHOAI)
  • Red Hat OpenShift AI 2.25
  • Red Hat OpenShift AI 3.3

Timeline

  1. 2026-10-02 Threatist record created from CVE-2026-22778
  2. 2026-10-03 Alias added: GHSA-4r2x-xpjr-7cvv
  3. 2026-10-12 Priority changed: MEDIUM → HIGH

References

Patch 2

Advisory 3

Other 15

  • TI-2024-10229 Remote Code Execution in vllm-project/vllm CVSS 3.0CRITICAL 9.8
  • TI-2024-08404 Remote Code Execution in vllm-project/vllm CVSS 3.1CRITICAL 9.8
  • TI-2026-36017 vLLM: OpenAI auth bypass CVSS 3.1CRITICAL 9.1
  • TI-2026-63056 vLLM before 0.28.0 Denial of Service via negative token ID CVSS 4.0HIGH 8.7
  • TI-2026-04308 Hardcoded trust_remote_code=True in vllm-project/vllm Bypasses User Security Control CVSS 3.0HIGH 8.8
  • TI-2026-39137 vLLM: Remote DoS in vLLM via Invalid Recovered Token Reinjection CVSS 3.1HIGH 7.5
See all 124 →

Sources

This product uses the NVD API but is not endorsed or certified by the NVD. Sources and licences