Read the docs
TI-2026-51647 Aliases CVE-2026-73557GHSA-pr7f-p5mw-fc87

vLLM: Incomplete CVE-2025-62164 remediation can be bypassed by concurrent prompt parts

Priority LOW 28CWE-362Published · Updated

Key facts

Attack vector
Network
Privileges required
None
User interaction
None
Attack complexity
Low
Attack requirements
Present
Impact type
—
Confidentiality · integrity · availability
None · None · Low
Fix status
Fixed

Attack profile from CVSS 4.0 · CNA. Fix status from CNA, GitHub and Red Hat.

Summary

vLLM is an inference and serving engine for large language models. From 0.20.2rc0 until 0.26.0, safe_load_prompt_embeds in vllm/renderers/embed_utils.py uses torch.sparse.check_sparse_tensor_invariants, whose process-global save, enable, and restore state can be raced by concurrent prompt_embeds parts submitted to POST /v1/chat/completions through AsyncMultiModalItemTracker.resolve_items, asyncio.gather, and the default executor, allowing an invalid sparse tensor to reach tensor.to_dense despite the CVE-2025-62164 guard when enable_prompt_embeds is enabled. This issue is fixed in version 0.26.0.

Why it matters now

Exploitation None reported CISA SSVC · not in CISA KEV
EPSS 0.00247 16th percentileScored 2026-11-07
Fix Fixed vllm (vllm-project): 0.26.0vllm (PyPI): 0.26.0Red Hat: 7 advisories

SSVC decision points

CISA ADP · SSVC 2.0.3

Exploitation
None
Automatable
No
Technical impact
Partial

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-projectvllm0.20.2rc0 to before 0.26.00.26.0Patch availableCNA
—vllm (PyPI)0.21.0 to before 0.26.00.26.0Patch availableGHSA-pr7f-p5mw-fc87
  • vllm Vendorvllm-project Affected 0.20.2rc0 to before 0.26.0 Fixed in 0.26.0 Patch available CNA
  • vllm (PyPI) Vendor— Affected 0.21.0 to before 0.26.0 Fixed in 0.26.0 Patch available GHSA-pr7f-p5mw-fc87

CPE configurations NVD · 1

CPEVersionsRoleVendor
cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*0.21.0 to before 0.26.0VulnerablevLLM
  • cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:* Versions 0.21.0 to before 0.26.0 Vulnerable ·vLLM

Remediation

  • Fix · CNA vllm (vllm-project) Fixed in 0.26.0
  • Fix · GHSA-pr7f-p5mw-fc87 vllm (PyPI) Fixed in 0.26.0 Advisory →

Red Hat 7 advisories

  • RHSA-2026:92452 Released 2026-10-29 · Red Hat rating Critical
    Red Hat Enterprise Linux AI 3.5.2 enhancement update
    • Red Hat Enterprise Linux AI 3.5 registry.redhat.io/rhelai3/bootc-aws-cuda-rhel9, registry.redhat.io/rhelai3/bootc-azure-cuda-rhel9, registry.redhat.io/rhelai3/bootc-azure-rocm-rhel9 +3
  • RHSA-2026:92449 Released 2026-10-29 · Red Hat rating Critical
    Red Hat Enterprise Linux AI 3.5.2 enhancement update
    • Red Hat Enterprise Linux AI 3.5 registry.redhat.io/rhelai3/disk-image-cuda-rhel9
  • RHSA-2026:82415 Released 2026-10-13 · Red Hat rating Critical
    Red Hat AI Inference 3.5.2 (cpu)
    • Red Hat AI Inference Server 3.5 registry.redhat.io/rhaii/vllm-cpu-rhel9
Show 4 more Red Hat advisories
  • RHSA-2026:82371 Released 2026-10-13 · Red Hat rating Critical
    Red Hat AI Inference 3.5.2 (spyre)
    • Red Hat AI Inference Server 3.5 registry.redhat.io/rhaii/vllm-spyre-rhel9
  • RHSA-2026:82368 Released 2026-10-13 · Red Hat rating Critical
    Red Hat AI Inference 3.5.2 (cuda)
    • Red Hat AI Inference Server 3.5 registry.redhat.io/rhaii/vllm-cuda-rhel9
  • RHSA-2026:82061 Released 2026-10-13 · Red Hat rating Critical
    Red Hat AI Inference Tech Preview 3.5.2 (Gaudi)
    • Red Hat AI Inference Server 3.5 registry.redhat.io/rhaii/vllm-gaudi-rhel9
  • RHSA-2026:82059 Released 2026-10-13 · Red Hat rating Critical
    Red Hat AI Inference 3.5.2 (ROCm)
    • Red Hat AI Inference Server 3.5 registry.redhat.io/rhaii/vllm-rocm-rhel9

Vendor assessments

Red Hat

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

ProductStatusPlan or fix
Red Hat AI Inference Server 3.5Fixed RHSA-2026:82059, RHSA-2026:82061, RHSA-2026:82368, RHSA-2026:82371, RHSA-2026:82415
Red Hat Enterprise Linux AI 3.5Fixed RHSA-2026:92449, RHSA-2026:92452
Not affected: 3 products
  • Red Hat AI Inference Server
  • Red Hat Enterprise Linux AI (RHEL AI) 3
  • Red Hat OpenShift AI (RHOAI)

Timeline

  1. 2026-10-02 Threatist record created from CVE-2026-73557
  2. 2026-10-03 Alias added: GHSA-pr7f-p5mw-fc87
  3. 2026-10-13 Fix available: RHSA-2026:82061 · Red Hat
  4. 2026-10-13 Fix available: RHSA-2026:82059 · Red Hat
  5. 2026-10-13 Fix available: Red Hat AI Inference Server 3.5 · Red Hat
  6. 2026-10-14 Fix available: RHSA-2026:82368 · Red Hat
  7. 2026-10-14 Fix available: RHSA-2026:82415 · Red Hat
  8. 2026-10-14 Fix available: RHSA-2026:82371 · Red Hat
  9. 2026-10-15 Affected versions changed · Red Hat
  10. 2026-10-16 CVSS score added · NVD
  11. 2026-10-30 Fix available: RHSA-2026:92452 · Red Hat
  12. 2026-10-30 Fix available: RHSA-2026:92449 · Red Hat
  13. 2026-10-30 Fix available: Red Hat Enterprise Linux AI 3.5 · Red Hat

References

Patch 2

Advisory 2

Other 2

  • 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