Read the docs
TI-2025-29283 Aliases CVE-2025-46570GHSA-4qjh-9fv9-r85r

vLLM’s Chunk-Based Prefix Caching Vulnerable to Potential Timing Side-Channel

Priority LOW 14CWE-208Published · Updated

Key facts

Attack vector
Network
Privileges required
Low
User interaction
Required
Attack complexity
High
Impact type
Information disclosure
Confidentiality · integrity · availability
Low · None · None
Fix status
Partial Red Hat: 2 without a fix

Attack profile from CVSS 3.1 · CNA. Fix status from CNA, GitHub and Red Hat. Impact type from CWE-208.

Summary

vLLM is an inference and serving engine for large language models (LLMs). Prior to version 0.9.0, when a new prompt is processed, if the PageAttention mechanism finds a matching prefix chunk, the prefill process speeds up, which is reflected in the TTFT (Time to First Token). These timing differences caused by matching chunks are significant enough to be recognized and exploited. This issue has been patched in version 0.9.0.

Why it matters now

Exploitation None reported CISA SSVC · not in CISA KEV
EPSS 0.0043 37th percentileScored 2026-11-07
Fix Partial vllm (vllm-project): 0.9.0vllm (PyPI): 0.9.0Red Hat: 2 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 0.9.00.9.0Patch availableCNA
—vllm (PyPI)before 0.9.00.9.0Patch availableGHSA-4qjh-9fv9-r85r
  • vllm Vendorvllm-project Affected before 0.9.0 Fixed in 0.9.0 Patch available CNA
  • vllm (PyPI) Vendor— Affected before 0.9.0 Fixed in 0.9.0 Patch available GHSA-4qjh-9fv9-r85r

CPE configurations NVD · 1

CPEVersionsRoleVendor
cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*before 0.9.0VulnerablevLLM
  • cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:* Versions before 0.9.0 Vulnerable ·vLLM

Remediation

  • Fix · CNA vllm (vllm-project) Fixed in 0.9.0
  • Fix · GHSA-4qjh-9fv9-r85r vllm (PyPI) Fixed in 0.9.0 Advisory →

Vendor assessments

Red Hat

Red Hat rating: Low. 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)AffectedNo fix yet: Fix deferred

Timeline

  1. 2026-10-02 Threatist record created from CVE-2025-46570
  2. 2026-10-03 Alias added: GHSA-4qjh-9fv9-r85r

References

Patch 1

Advisory 3

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