2.6

CVE-2025-46570: vllm-project vllm Observable Timing Discrepancy

CVSS: 2.6v3.1LowAnalyzed
AliasesGHSA-4qjh-9fv9-r85rPYSEC-2025-53
Published 29-05-2025Last modified 29-05-2025

Vulnerability summary

CVE-2025-46570 is a low (CVSS 2.6) Observable Timing Discrepancy vulnerability in vllm-project vllm. EPSS estimates a 0.34% chance of exploitation in the next 30 days, higher than 25.5% of scored vulnerabilities.

Overall risk
10/100
0100
Low · default modelHow is this calculated?
CVSS score
2.6/10
010
Low · v3.1 (CVE.org)
EPSS score
0.34%
0%100%
Low · +0.04 pp in 30 days
KEVNot listedIn no KEV catalog
RansomwareNone recordedNo known use recorded
Public exploits0No public exploits recorded

Advisory lineage

Upstream 0 · Downstream 0 · Related 3

Timeline

  1. 29-05-2025
    PublishedVulnerability first disclosed.
  2. 29-05-2025
    Last modifiedVulnerability information updated.

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.

CVSS metrics

3 scores · 3 sources · 1 version

EPSS trend

Probability

0.34%

Chance of exploitation in the next 30 days

Percentile

26%

Versus all other CVEs

30-day change

+0.04 pp

From 0.29% on 20-09-2026

20-09-202620-10-2026
EPSS score
0.34%

Known exploited

Compare KEV catalogs

Not in any of the 3 tracked catalogs.

Public exploits

0 found

No public exploits indexed.

Weaknesses

CWE · root cause

Affected systems

PyPI1 product
  • vllm< 77073c77bc2006eb80ea6d5128f076f5e6c6f54f | < 0.9.0
Vllm1 product
Vllm-project1 product

3 vendors · 3 products

References

6 links