CVE-2025-46570

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

LOW 2.6

At a glance

Source
NVD
EPSS
EPSS 0.43 % · percentile 37 ()
State
PUBLISHED
Published
Updated
Assigner
GitHub_M
NVD status
Analyzed

Description

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.

Weaknesses (CWE)

  • CWE-208 — Observable Timing Discrepancy
  • CWE-203 — Observable Discrepancy

References

Scores

VersionScoreVectorScored by
CVSS 3.1LOW 2.6CVSS:3.1/AV:N/AC:H/PR:L/UI:R/S:U/C:L/I:N/A:NGitHub_M
CVSS 3.1LOW 2.6CVSS:3.1/AV:N/AC:H/PR:L/UI:R/S:U/C:L/I:N/A:NNVD · security-advisories@github.com (Secondary)

Advisories

European references

CERT-FR advisories

Affected configurations (NVD CPE)

Affected products

ProductVersions
vllm-project · vllmaffected < 0.9.0

Sources

SourceLicence
CVE List (CVEProject cvelistV5)CVE Program Terms of Use
EPSS (FIRST.org)FIRST — attribution requested
ENISA EUVDENISA — reproduction authorised, source acknowledged
CERT-FR (ANSSI)Licence Ouverte / Etalab 2.0
NVD (NIST) via fkie-cad/nvd-json-data-feedsPublic domain — this product uses data from the NVD API but is not endorsed or certified by the NVD