CVE-2026-25960

HIGH

CVSS v3.1: 7.1 · EPSS: 0.0044 (35.0 percentile)

NetworkNo user interactionSSRFVendor advisory ref

Source data as of:

At a glance

Severity
HIGH
CVSS
7.1 v3.1 · NVD
EPSS
0.0044 (35.0 percentile) · FIRST.org
CISA KEV
No
Type
SSRF · NVD CWE
Attack conditions (CVSS vector)
NetworkNo user interaction · Source: NVD Vector
Affected vendors
vllm
Published
2026-03-09 · Modified: 2026-06-30

CVSS / EPSS / KEV

CVSS v3.1 7.1 / 10 HIGH Source: NVD
EPSS 0.0044 35.0 percentile Source: FIRST.org
CISA KEV No Source: CISA

Source — CVSS: NVD · EPSS: FIRST.org · KEV: CISA. Data & Sources

Description

vLLM is an inference and serving engine for large language models (LLMs). The SSRF protection fix for CVE-2026-24779 add in 0.15.1 can be bypassed in the load_from_url_async method due to inconsistent URL parsing behavior between the validation layer and the actual HTTP client. The SSRF fix uses urllib3.util.parse_url() to validate and extract the hostname from user-provided URLs. However, load_from_url_async uses aiohttp for making the actual HTTP requests, and aiohttp internally uses the yarl library for URL parsing. This vulnerability in 0.17.0.

Record details

CVE ID
CVE-2026-25960
CVSS (v3.1)
7.1 (HIGH)
Vector
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:L
Exploitability subscore
2.8
Impact subscore
4.2
EPSS
0.0044 (35.0 percentile) — 2026-06-30
CISA KEV
No
Weakness (CWE)
CWE-918, CWE-474
Affected vendors
vllm
Affected configurations (CPE)
1
Published
2026-03-09
Modified
2026-06-30
Status
Modified

References

Reference URLs as listed by NVD, grouped by a mechanical match on the link's host/pattern. Labels describe the link type only.