| Name | CVE-2026-44223 |
| Description | 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. |
| Source | CVE (at NVD; CERT, ENISA, LWN, oss-sec, fulldisc, Debian ELTS, Red Hat, Ubuntu, Gentoo, SUSE bugzilla/CVE, GitHub advisories/code/issues, web search, more) |
| Debian Bugs | 1095237 |
The information below is based on the following data on fixed versions.