Read the docs

Request ID reuse lets vLLM scoring mix queries between requests

Priority LOW 19CWE-639Published · Updated

Key facts

Attack vector
Network
Privileges required
Low
User interaction
None
Attack complexity
High
Impact type
Security bypass
Confidentiality · integrity · availability
None · Low · Low
Fix status
Partial Red Hat: 3 without a fix

Attack profile from CVSS 3.1 · CNA. Fix status from CNA, GitHub and Red Hat. Impact type from CWE-639.

Summary

vLLM versions before 0.30.0 use the caller-controlled X-Request-Id header to key cached query embeddings in flash late-interaction scoring at the /score and /rerank endpoints. A concurrent request that reuses another request's identifier can overwrite the cached query embedding (CWE-639), so the victim's documents are scored against the attacker's query. A remote attacker with low privileges can also trigger a late-interaction cache-miss error through shared use counters.

Written with Claude Haiku 5.5 from the sources below · 2026-10-29

Description from the CNA

vLLM is an inference and serving engine for large language models. Prior to 0.30.0, flash late-interaction scoring at the /score and /rerank endpoints derives each worker's query_key value from the caller-controlled X-Request-Id header. A concurrent request that reuses a victim's identifier can overwrite the cached query embedding so the victim's documents are scored against the attacker's query, and shared use counters can also cause a late-interaction cache-miss error. This issue is fixed in version 0.30.0.

Why it matters now

Exploitation None reported CISA SSVC · not in CISA KEV
EPSS 0.00202 10th percentileScored 2026-11-07
Fix Partial vllm (vllm-project): 030.0vllm (PyPI): 0.30.0Red Hat: 3 without a fix

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-projectvllmbefore 030.0030.0Patch availableCNA
—vllm (PyPI)before 0.30.00.30.0Patch availableGHSA-2phq-3phc-84px
  • vllm Vendorvllm-project Affected before 030.0 Fixed in 030.0 Patch available CNA
  • vllm (PyPI) Vendor— Affected before 0.30.0 Fixed in 0.30.0 Patch available GHSA-2phq-3phc-84px

CPE configurations NVD · 1

CPEVersionsRoleVendor
cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*before 0.30.0VulnerablevLLM
  • cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:* Versions before 0.30.0 Vulnerable ·vLLM

Remediation

Upgrade the vllm PyPI package to 0.30.0, which fixes the issue. Red Hat AI Inference Server, Red Hat Enterprise Linux AI (RHEL AI) 3 and Red Hat OpenShift AI (RHOAI) are affected. The sources list no fixed version for them.

Written with Claude Haiku 5.5 from the sources below · 2026-10-29

  • Fix · CNA vllm (vllm-project) Fixed in 030.0
  • Fix · GHSA-2phq-3phc-84px vllm (PyPI) Fixed in 0.30.0 Advisory →

Vendor assessments

Red Hat

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

ProductStatusPlan or fix
Red Hat AI Inference ServerAffectedNo fix yet: Fix deferred
Red Hat Enterprise Linux AI (RHEL AI) 3AffectedNo fix yet: Fix deferred
Red Hat OpenShift AI (RHOAI)AffectedNo fix yet: Fix deferred

Timeline

  1. 2026-10-06 Threatist record created from CVE-2026-105755
  2. 2026-10-06 Alias added: GHSA-2phq-3phc-84px
  3. 2026-10-06 Fix available: PyPI/[email protected] · GHSA-2phq-3phc-84px
  4. 2026-10-06 CVSS score added · GHSA-2phq-3phc-84px
  5. 2026-10-06 Weakness changed · GHSA-2phq-3phc-84px
  6. 2026-10-06 5 new references · GHSA-2phq-3phc-84px
  7. 2026-10-06 Affected versions changed · Red Hat
  8. 2026-10-06 CVSS score added · Red Hat
  9. 2026-10-06 CISA SSVC decision points changed

References

  • 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