Cyber Resilience

CVE-2025-46570

Vllm ≤ 0.9.0

Published
29 May 2025
Modified
17 June 2026
Patch / advisory
View advisory
CVSS Score v3.1 2.6
Click a component to see what it means
Raw vectorCVSS:3.1/AV:N/AC:H/PR:L/UI:R/S:U/C:L/I:N/A:N
EPSS Score 0.0026 17th percentile
Risk Priority 24 floored blend · peak EPSS

Summary

CVE-2025-46570 is a low-severity Observable Timing Discrepancy (CWE-208) vulnerability in Vllm Vllm. Its CVSS base score is 2.6 (Low).

Operationally, exploitation aligns with the MITRE ATT&CK technique Password Guessing (T1110.001); ranked at the 17th percentile by exploit likelihood (below the median); it is not currently listed in the CISA KEV catalog.

This vulnerability is AI-related — categorised as NLP and Transformers; in the Adversarial Attacks risk domain.

The strongest mitigations our analysis identified map to IA-6 (Authentication Feedback) and SA-11 (Developer Testing and Evaluation) — see the control section below for these in your framework.

EU & UK References

Vulnerability Data

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…

more

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.

CWE(s)
CWE-208CWE-203

AI Security AnalysisAI

AI Category
NLP and Transformers
Risk Domain
Adversarial Attacks
OWASP Top 10 for LLMs 2025
None mapped
Classification Reason
Matched keywords: llms, vllm

Related Threats

MITRE ATT&CK Enterprise Techniques

T1110.001 Password Guessing Credential Access
Adversaries with no prior knowledge of legitimate credentials within the system or environment may guess passwords to attempt access to accounts.
attack.mitre.org →
T1087 Account Discovery Discovery
Adversaries may attempt to get a listing of valid accounts, usernames, or email addresses on a system or within a compromised environment.
attack.mitre.org →
T1087.001 Local Account Discovery
Adversaries may attempt to get a listing of local system accounts.
attack.mitre.org →
T1087.002 Domain Account Discovery
Adversaries may attempt to get a listing of domain accounts.
attack.mitre.org →
T1087.003 Email Account Discovery
Adversaries may attempt to get a listing of email addresses and accounts.
attack.mitre.org →
T1087.004 Cloud Account Discovery
Adversaries may attempt to get a listing of cloud accounts.
attack.mitre.org →
Derived from this CVE’s CWE(s) via the direct CWE→ATT&CK cross-walk.

CVEs Like This One

CVE-2026-22778Same product: Vllm Vllm
CVE-2024-11041Same product: Vllm Vllm
CVE-2026-34755Same product: Vllm Vllm
CVE-2025-30202Same product: Vllm Vllm
CVE-2026-34760Same product: Vllm Vllm
CVE-2025-46560Same product: Vllm Vllm
CVE-2025-29770Same product: Vllm Vllm
CVE-2025-48942Same product: Vllm Vllm
CVE-2025-25183Same product: Vllm Vllm
CVE-2025-62426Same product: Vllm Vllm

Affected Assets

vllm
vllm
≤ 0.9.0

Mitigating Controls

Control response

Prevent
Stop it (NIST 800-53)

—

Detect
Catch it (NIST detect / respond)

—

Harden
Shrink the surface (DISA STIG)
  • 1 hardening rule · 1 OS baseline
Validate
Prove the fix (OWASP ASVS)
  • V11.2.4

Mitigating Controls (NIST 800-53 r5) AI

IA-6 Authentication Feedback good match

Obscures authentication feedback so that success/failure differences are not observable to attackers.

SA-11 Developer Testing and Evaluation good match

Developer testing can include timing analysis or side-channel test cases that reveal observable timing discrepancies.

SI-11 Error Handling good match

Requires error messages to avoid revealing exploitable details, directly stopping observable response discrepancies.

SA-8 Security and Privacy Engineering Principles partial match

Engineering principles can mandate constant-time algorithms and side-channel resistance so timing discrepancies are never introduced.

SC-13 Cryptographic Protection partial match

Requiring approved cryptographic modules and algorithms implicitly demands implementations free of observable timing leaks.

Mitigating Controls (NIST CSF 2.0) AI

Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→CSF cross-walk (authority under review) — links open the control.

PR.PS-06 mostly match
prevents

Secure SDLC practices directly require constant-time implementations that eliminate observable timing discrepancies.

Mitigating Controls (ISO/IEC 27001:2022 Annex A) AI

Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→ISO cross-walk (authority under review) — links open the control.

A.8.17 Clock synchronization none match
none

Consistent reference clocks limit the attacker's ability to measure or manipulate timing differences that could reveal internal state or processing paths.

References