{"id":"CVE-2026-94625","title":"vLLM through 0.29.0 contains a resource exhaustion vulnerability in MooncakeConnector where rejected prefill requests create ownerless transfer placeholders that are never reclaimed","summary":"vLLM through 0.29.0 contains a resource exhaustion vulnerability in MooncakeConnector where rejected prefill requests create ownerless transfer placeholders that are never reclaimed. Attackers can send rejected requests to exhaust sender…","severity":"medium","cvss":5.3,"cvssVector":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L","cwe":["CWE-772"],"vendor":"vllm-project","product":"vllm","affected":["vllm <= 0.29.0"],"published":"2026-09-21","updated":"2026-09-23","sourceUpdated":"2026-09-23T15:17:30.077","source":"NVD","sourceUrl":"https://nvd.nist.gov/vuln/detail/CVE-2026-94625","references":[{"url":"https://github.com/vllm-project/vllm","label":"disclosure@vulncheck.com"},{"url":"https://github.com/vllm-project/vllm/blob/v0.29.0/vllm/distributed/kv_transfer/kv_connector/v1/mooncake/mooncake_connector.py#L1234-L1242","label":"disclosure@vulncheck.com"},{"url":"https://github.com/vllm-project/vllm/pull/51236","label":"disclosure@vulncheck.com"},{"url":"https://www.vulncheck.com/advisories/vllm-through-0.29.0-resource-exhaustion-via-ownerless-mooncake-transfer-placeholders","label":"disclosure@vulncheck.com"},{"url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-94625.json"},{"url":"https://access.redhat.com/security/cve/CVE-2026-94625"},{"url":"https://bugzilla.redhat.com/show_bug.cgi?id=2538868"},{"url":"https://www.cve.org/CVERecord?id=CVE-2026-94625"},{"url":"https://nvd.nist.gov/vuln/detail/CVE-2026-94625"}],"tags":["nvd","cve.org","csaf","vex","red-hat","score-dispute"],"ssvc":{"exploitation":"none","automatable":"yes","technicalImpact":"partial","timestamp":"2026-09-23T15:09:07.861181Z"},"epss":0.00524,"epssPercentile":0.41917,"scores":{"nvd":5.3,"vendor":7.5},"ingestedAt":"2026-09-21T22:54:37.973Z","slug":"CVE-2026-94625","body":"## Overview\n\nvLLM through 0.29.0 contains a resource exhaustion vulnerability in MooncakeConnector where rejected prefill requests create ownerless transfer placeholders that are never reclaimed. Attackers can send rejected requests to exhaust sender task pools, causing valid requests to be delayed by up to 480 seconds while health checks continue returning success.\n\n## Remediation\n\nRefer to the linked advisories for vendor-supplied fixes and affected version ranges.\n\n## Vendor advisories\n\n- **Red Hat VEX** · Important · affected: Red Hat AI Inference Server, Red Hat Enterprise Linux AI (RHEL AI) 3, Red Hat OpenShift AI (RHOAI) · no fix planned: Red Hat AI Inference Server, Red Hat Enterprise Linux AI (RHEL AI) 3, Red Hat OpenShift AI (RHOAI) · updated 2026-09-24 · [vex](https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-94625.json)","depth":"sunlit","depthScore":29,"depthScoreParts":{"impact":29.2,"likelihood":0.1,"exploitation":0,"ransomware":0},"changes":[]}