{"cve":"CVE-2026-44223","nvd":{"published":"2026-05-12T20:16:43.293","modified":"2026-06-22T22:16:45.507","status":"Modified","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.","cvss_v4":null,"cvss_v3":{"version":"3.1","vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H","score":6.5},"cwes":["CWE-131","CWE-704"],"references":["https://github.com/vllm-project/vllm/pull/38610","https://github.com/vllm-project/vllm/security/advisories/GHSA-83vm-p52w-f9pw","https://github.com/vllm-project/vllm/pull/38610","https://github.com/vllm-project/vllm/security/advisories/GHSA-83vm-p52w-f9pw"],"products":["vllm:vllm"],"source":"nvd","synced_at":"2026-10-19T22:35:16.954578616+00:00"},"epss":{"cve":"CVE-2026-44223","score":0.00367,"percentile":0.30561,"date":"2026-11-01"},"kev":null,"osv":[{"id":"GHSA-83vm-p52w-f9pw","url":"https://osv.dev/vulnerability/GHSA-83vm-p52w-f9pw","summary":"vLLM: extract_hidden_states speculative decoding crashes server on any request with penalty parameters","modified":"2026-09-10T03:50:46.257654637Z","packages":[{"ecosystem":"PyPI","name":"vllm","purl":"pkg:pypi/vllm","ranges":[">=0.18.0 <0.20.0"],"fixed":["0.20.0"]}]},{"id":"PYSEC-2026-145","url":"https://osv.dev/vulnerability/PYSEC-2026-145","summary":"","modified":"2026-05-20T09:19:21.596358Z","packages":[{"ecosystem":"PyPI","name":"vllm","purl":"pkg:pypi/vllm","ranges":[">=0.18.0 <0.20.0"],"fixed":["0.20.0"]}]}],"attribution":["This product uses the NVD API but is not endorsed or certified by the NVD.","EPSS scores: FIRST Exploit Prediction Scoring System, https://www.first.org/epss","Known exploited vulnerabilities: CISA Known Exploited Vulnerabilities Catalog, https://www.cisa.gov/known-exploited-vulnerabilities-catalog","Package advisories: OSV (https://osv.dev), from the GitHub Advisory Database, PyPA, Go, RustSec and other ecosystem databases, under CC-BY 4.0 or CC0","CVE records: the CVE Program's CVE List (https://www.cve.org), under its terms of use (https://www.cve.org/Legal/TermsOfUse). CVE is a registered trademark of The MITRE Corporation; this product is not affiliated with the CVE Program."]}