CVE-2026-44223
Gravedad CVSS v3.1:
MEDIA
Tipo:
No Disponible / Otro tipo
Fecha de publicación:
12/05/2026
Última modificación:
12/05/2026
Descripción
*** Pendiente de traducción *** vLLM is an inference and serving engine for large language models (LLMs). From 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.
Impacto
Puntuación base 3.x
6.50
Gravedad 3.x
MEDIA