SOC status:Duty analyst on shift

UK Cyber Defence
VulnerabilityAnalyzed

CVE-2026-105754

Forged grid geometry, field types, or non-positive placeholder lengths can terminate the shared EngineCore; when an attacker knows or can induce a victim's content hash, forged cache hashes can poison or retrieve cross-request encoder-cache state; and…

MEDIUM 6.5EPSS 0.27%

Does this matter?

Lower severity and a low EPSS score (0.27%). Track it; it rarely justifies an emergency change on its own.

Description

vLLM is an inference and serving engine for large language models. Prior to 0.30.0, the /inference/v1/generate endpoint in the disaggregated scale-out path accepts caller-supplied tensors in the features.kwargs_data field, cache identifiers in the features.mm_hashes field, ranges in the features.mm_placeholders field, and wire-selected multimodal field processors without rebinding them to the active model renderer contract. Forged grid geometry, field types, or non-positive placeholder lengths can terminate the shared EngineCore; when an attacker knows or can induce a victim's content hash, forged cache hashes can poison or retrieve cross-request encoder-cache state; and dropped sparse placeholder masks can alter replayed transport semantics. This issue is fixed in version 0.30.0.

CVSS 3.1
6.5 MEDIUMCVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H
EPSS
0.27% probability · 19th percentile
Public exploits
None in Exploit-DB
CISA KEV
Not listed
Weakness
CWE-20, CWE-617, CWE-639, CWE-668, CWE-704, CWE-1284
Affected
vllm/vllm
Source
security-advisories@github.com
EPSS trend
0%3%5%6 October 2026: 0.27%26 October 2026: 0.27%6 October 202626 October 2026

EPSS moved from 0.3% on 6 October 2026 to 0.3% on 26 October 2026.

Source: NVD record, EPSS from FIRST.org, KEV from CISA, exploits from Exploit-DB. Refreshed daily. Download this record as JSON.