{"id":"CVE-2026-114049","published":"2026-10-21T01:17:27.513","lastModified":"2026-10-22T15:44:59.783","description":"vLLM before 0.31.0 contains a cross-request mask leak in InputBatch.condense() that allows remote attackers to constrain other clients' outputs by leaving stale allowed_token_ids rows. Attackers can submit concurrent constrained generate requests, including via unauthenticated /invocations, so recycled batch slots apply their token allowlist to victims' unrestricted requests.","cvssScore":3.7,"cvssSeverity":"LOW","cvssVector":"CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:L/A:N","cwes":["CWE-440"],"vendors":[],"products":[],"references":[{"url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-6cxc-2vcg-w5qc","tags":[]},{"url":"https://www.vulncheck.com/advisories/vllm-before-0.31.0-stale-allowed-token-ids-mask-via-inputbatch-condense","tags":[]}],"exploitRefs":[{"url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-6cxc-2vcg-w5qc","tags":[]}],"hasPoc":true,"ai":null}