{"id":"CVE-2026-44223","aliases":["GHSA-83vm-p52w-f9pw","PYSEC-2026-145"],"url":"https://o3.security/vulnerability/CVE-2026-44223","summary":"vLLM: extract_hidden_states speculative decoding crashes server on any request with penalty parameters","details":"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.","published":"2026-05-12T19:58:40.862Z","modified":"2026-08-07T11:31:17.218975732Z","cvss":{"score":6.5,"severity":"MEDIUM","vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H"},"epss":null,"cisaKev":null,"exploitsKnown":0,"affectedPackages":[{"ecosystem":"PyPI","name":"vllm","fixedVersion":"0.20.0"}],"fix":{"url":"https://github.com/vllm-project/vllm/pull/38610","label":"vllm-project/vllm#38610"},"references":[{"type":"ADVISORY","url":"https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/44xxx/CVE-2026-44223.json"},{"type":"ADVISORY","url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-83vm-p52w-f9pw"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-44223"},{"type":"FIX","url":"https://github.com/vllm-project/vllm/pull/38610"}],"provenance":{"sources":["OSV.dev","FIRST.org (EPSS)"],"lastVerified":"2026-08-07T11:31:17.218975732Z"}}