Detections
Analytics

Chainguard: vllm-openai-cuda-12.9: security update to 0.28.0-r0

medium Tenable Self-Hosted Container Security Plugin ID 476422

Description

There are packages installed that are affected by a vulnerability referenced in the following CVE:

 - vLLM is an inference and serving engine for large language models (LLMs). From 0.18.0 to before 0.20.0,
 the extract_hidden_states speculative decoding proposer in vLLM returns a tensor with an incorrect shape
 after the first decode step, causing a RuntimeError that crashes the EngineCore process. The crash is
 triggered when any request in the batch uses sampling penalty parameters (repetition_penalty,
 frequency_penalty, or presence_penalty). A single request with a penalty parameter (e.g.,
 "repetition_penalty": 1.1) is sufficient to crash the server. This vulnerability is fixed in 0.20.0.
 (CVE-2026-44223)

Solution

Update the vllm-openai-cuda-12.9 library and its related packages to version 0.28.0-r0 or later.

Plugin Details

Severity: Medium

ID: 476422

Version: Revision 1.1

Type: Local

Published: 10/21/2026

Updated: 10/21/2026

Risk Information

VPR

Risk Factor: Low

Score: 3

Percentile: 23.67

CVSS v2

Risk Factor: Medium

Base Score: 6.8

Temporal Score: 5

Vector: CVSS2#AV:N/AC:L/Au:S/C:N/I:N/A:C

CVSS Score Source: CVE-2026-44223

CVSS v3

Risk Factor: Medium

Base Score: 6.5

Temporal Score: 5.7

Vector: CVSS:3.0/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H

Temporal Vector: CVSS:3.0/E:U/RL:O/RC:C

Vulnerability Information

Exploit Ease: No known exploits are available

Vulnerability Publication Date: 5/6/2026

Reference Information

CVE: CVE-2026-44223