Not a U.S. government website. NDD is an independent vulnerability database by Volerion and is not affiliated with or endorsed by NIST or NVD.
VOLERION
Volerion Security Research

NOT DEFERRED DATABASE

VULNERABILITIES

CVE-2026-44223 Details

Description

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.

Metrics

CVSS 3.x Severity and Vector Strings:

References to Advisories, Solutions, and Tools

By selecting these links, you will be leaving this site. These are references gathered from the official CVE record and are not endorsed by Volerion.

Weakness Enumeration

CWE-IDCWE NameSource
CWE-131Incorrect Calculation of Buffer Size[email protected]
CWE-704Incorrect Type Conversion or Cast[email protected]

Affected Products

ProductVersions
vllm vllm
>= 0.18.0, < 0.20.0

CPE

  • cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*

Remediation

  • No remediation found in references.

Change History

7 change records found show changes


QUICK INFO

CVE Dictionary Entry:
CVE-2026-44223
NVD Published Date:
May 12, 2026
NVD Last Modified:
Nov 11, 2026
Source:
[email protected]
CVE-2026-44223 Details - Not Deferred