CVE-2025-46570 | vLLM’s Chunk-Based Prefix Caching Vulnerable to Potential Timing Side-Channel

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.

Published: 2025-05-29 Last update: 2025-06-24 Assigner: [email protected] Source: [email protected]
NVD Status：Analyzed，CVE State: published
View at NVD, CVE.org, EUVD

CVE-2025-46570 is rated Low Risk (24.2/100): CVSS Low severity, with low exploitation likelihood (EPSS 0.18%). Monitor for updates and reassess as exploit intelligence or EPSS changes.

Risk is dynamic; we continuously reassess and refresh what is shown on this page as upstream context changes.

Exploit prediction scoring system (EPSS) score for CVE-2025-46570

EPSS lead: Daily EPSS estimates relative likelihood of exploitation; percentile ranks this CVE among scored vulnerabilities (higher = more severe relative rank).

# Date Old EPSS score New EPSS score Delta (New - Old)
1 2026-03-25 0.05% 0.18% +0.13%
2 2026-02-12 0.03% 0.05% +0.02%
3 2025-11-21 — 0.03% —

Full EPSS history (5 records total)

Common vulnerability scoring system (CVSS) metrics for CVE-2025-46570

CVSS metrics for this CVE.

Base score Version Severity Vector Exploitability Impact Score source
2.6 3.1 LOW
CVSS:3.1/AV:N/AC:H/PR:L/UI:R/S:U/C:L/I:N/A:N Click to expand
Attack vector (AV:N)
Could be attacked over the internet or any normal routed network—not just someone sitting at the machine.
Attack complexity (AC:H)
Even with access, the exploit needs extra luck, timing, or a fussy environment to actually work.
Privileges required (PR:L)
A normal user session is enough; they don’t have to be admin.
User interaction (UI:R)
A real person has to do something—click, install, enable—otherwise it doesn’t land.
Scope (S:U)
Damage stays in the same “trust bubble” as the broken component—no big spill into unrelated systems.
Confidentiality (C:L)
Some sensitive info could get out, but not a total data dump.
Integrity (I:N)
Data isn’t meaningfully altered or forged.
Availability (A:N)
Service keeps running; no real outage angle.
 1.2 1.4 [email protected]

Weakness enumeration for CVE-2025-46570

GitHub Security Advisory for CVE-2025-46570

GHSA-4qjh-9fv9-r85r · Severity: low · Ecosystem: pip — Potential Timing Side-Channel Vulnerability in vLLM’s Chunk-Based Prefix Caching

OS Trackers for CVE-2025-46570

vendor priority summary link
redhat low — https://access.redhat.com/security/cve/CVE-2025-46570

Affected software / configurations for CVE-2025-46570

Vendor Product Version Raw CPE
vllm vllm < 0.9.0 cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*

References for CVE-2025-46570

cvelogic Threat Intelligence