{"id":"CVE-2026-44223","title":"vLLM is an inference and serving engine for large language models (LLMs)","summary":"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,…","severity":"medium","cvss":6.5,"cvssVector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H","cwe":["CWE-131","CWE-704","CWE-130"],"published":"2026-05-12","updated":"2026-06-22","source":"NVD","sourceUrl":"https://nvd.nist.gov/vuln/detail/CVE-2026-44223","references":[{"url":"https://github.com/vllm-project/vllm/pull/38610","label":"security-advisories@github.com"},{"url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-83vm-p52w-f9pw","label":"security-advisories@github.com"},{"url":"https://github.com/vllm-project/vllm/pull/38610","label":"134c704f-9b21-4f2e-91b3-4a467353bcc0"},{"url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-83vm-p52w-f9pw","label":"134c704f-9b21-4f2e-91b3-4a467353bcc0"},{"url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-44223.json"},{"url":"https://access.redhat.com/security/cve/CVE-2026-44223"},{"url":"https://bugzilla.redhat.com/show_bug.cgi?id=2476827"},{"url":"https://www.cve.org/CVERecord?id=CVE-2026-44223"},{"url":"https://nvd.nist.gov/vuln/detail/CVE-2026-44223"},{"url":"https://access.redhat.com/errata/RHSA-2026:57380"},{"url":"https://access.redhat.com/errata/RHSA-2026:57389"},{"url":"https://access.redhat.com/errata/RHSA-2026:57390"},{"url":"https://access.redhat.com/errata/RHSA-2026:57387"},{"url":"https://access.redhat.com/errata/RHSA-2026:59151"},{"url":"https://access.redhat.com/errata/RHSA-2026:59144"},{"url":"https://github.com/pypa/advisory-database/tree/main/vulns/vllm/PYSEC-2026-145.yaml"},{"url":"https://github.com/vllm-project/vllm"},{"url":"https://access.redhat.com/errata/RHSA-2026:69466"},{"url":"https://access.redhat.com/errata/RHSA-2026:69467"},{"url":"https://access.redhat.com/errata/RHSA-2026:69469"},{"url":"https://access.redhat.com/errata/RHSA-2026:69464"}],"tags":["nvd","csaf","vex","red-hat","osv","pip"],"epss":0.00367,"epssPercentile":0.30379,"ingestedAt":"2026-06-29T13:24:34.843Z","vendor":"Red Hat","product":"Red Hat Enterprise Linux AI 3.4","affected":["ai_inference_server","openshift_ai_rhoai","ai_inference_server 3.4","enterprise_linux_ai 3.4"],"patched":["ai_inference_server 3.4","enterprise_linux_ai 3.4"],"scores":{"nvd":6.5,"vendor":7.5,"osv":6.5},"aliases":["GHSA-83vm-p52w-f9pw","PYSEC-2026-145"],"ecosystem":"pip","slug":"CVE-2026-44223","body":"## Overview\n\nvLLM 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.\n\n## Remediation\n\nRefer to the linked advisories for vendor-supplied fixes and affected version ranges.\n\n## Vendor advisories\n\n- **RHSA-2026:57380** · Red Hat · fixed in: Red Hat AI Inference Server 3.4 · released 2026-08-20 · [advisory](https://access.redhat.com/errata/RHSA-2026:57380)\n- **RHSA-2026:57389** · Red Hat · fixed in: Red Hat AI Inference Server 3.4 · released 2026-08-20 · [advisory](https://access.redhat.com/errata/RHSA-2026:57389)\n- **RHSA-2026:57390** · Red Hat · fixed in: Red Hat AI Inference Server 3.4 · released 2026-08-20 · [advisory](https://access.redhat.com/errata/RHSA-2026:57390)\n- **RHSA-2026:57387** · Red Hat · fixed in: Red Hat AI Inference Server 3.4 · released 2026-08-20 · [advisory](https://access.redhat.com/errata/RHSA-2026:57387)\n- **RHSA-2026:59151** · Red Hat · fixed in: Red Hat Enterprise Linux AI 3.4 · released 2026-08-24 · [advisory](https://access.redhat.com/errata/RHSA-2026:59151)\n- **RHSA-2026:59144** · Red Hat · fixed in: Red Hat Enterprise Linux AI 3.4 · released 2026-08-24 · [advisory](https://access.redhat.com/errata/RHSA-2026:59144)\n- **Red Hat VEX** · Important · affected: Red Hat AI Inference Server, Red Hat OpenShift AI (RHOAI) · no fix planned: Red Hat AI Inference Server, Red Hat OpenShift AI (RHOAI) · updated 2026-09-21 · [vex](https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-44223.json)\n- **RHSA-2026:69466** · Red Hat · fixed in: Red Hat AI Inference Server 3.4 · released 2026-09-21 · [advisory](https://access.redhat.com/errata/RHSA-2026:69466)\n- **RHSA-2026:69467** · Red Hat · fixed in: Red Hat AI Inference Server 3.4 · released 2026-09-21 · [advisory](https://access.redhat.com/errata/RHSA-2026:69467)\n- **RHSA-2026:69469** · Red Hat · fixed in: Red Hat AI Inference Server 3.4 · released 2026-09-21 · [advisory](https://access.redhat.com/errata/RHSA-2026:69469)\n- **RHSA-2026:69464** · Red Hat · fixed in: Red Hat AI Inference Server 3.4 · released 2026-09-21 · [advisory](https://access.redhat.com/errata/RHSA-2026:69464)\n\n## Package advisory (CVE-2026-44223)\n\nAffected packages:\n\n- `vllm >= 0.18.0, < 0.20.0`\n\nPatched in:\n\n- `vllm 0.20.0`\n\nSource: https://osv.dev/vulnerability/GHSA-83vm-p52w-f9pw","depth":"sunlit","depthScore":36,"depthScoreParts":{"impact":35.8,"likelihood":0.1,"exploitation":0,"ransomware":0},"changes":[]}