{"id":"CVE-2026-114051","aliases":[],"cve":{"records":[{"id":"CVE-2026-114051","source":"VulnCheck","state":"PUBLISHED","reasons":[],"replacements":[],"disputed":false,"unsupportedWhenAssigned":false,"reserved":"2026-10-20T21:42:27.669Z","published":"2026-10-21T00:36:28.373Z","modified":"2026-10-28T15:15:55.536Z"}],"publishers":[{"name":"VulnCheck","role":"CNA","organizationId":"83251b91-4cc7-4094-a5c7-464a1b83ea10","modified":"2026-10-21T00:36:28.373Z"},{"name":"CISA-ADP","role":"ADP","organizationId":"134c704f-9b21-4f2e-91b3-4a467353bcc0","modified":"2026-10-28T15:15:55.536Z"}],"references":[{"url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-4xqp-c3mv-qff7","tags":["vendor-advisory"],"source":"VulnCheck"},{"url":"https://www.vulncheck.com/advisories/vllm-before-0.31.0-denial-of-service-via-min-tokens-in-inference-v1-generate","tags":["third-party-advisory"],"source":"VulnCheck"}],"weaknesses":[{"id":"CWE-248","description":"Uncaught Exception","source":"VulnCheck"}],"applicability":[{"source":"VulnCheck","statements":[{"nodes":[{"operator":"OR","negate":false,"cpeMatch":[{"vulnerable":true,"criteria":"cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*","versionEndExcluding":"0.31.0"}]}]}]}],"prioritization":[{"source":"CISA-ADP","type":"SSVC","role":"CISA Coordinator","timestamp":"2026-10-28T15:08:13.315794Z","decisions":[{"name":"Exploitation","value":"none"},{"name":"Automatable","value":"yes"},{"name":"Technical Impact","value":"partial"}]}]},"title":"vLLM before 0.31.0 Denial of Service via min_tokens in /inference/v1/generate","titleSource":"VulnCheck","descriptionSource":"VulnCheck","description":"vLLM before 0.31.0 skips the min_tokens <= max_tokens check for /inference/v1/generate requests with a null max_tokens, so the EngineCore input thread later dies on an uncaught VLLMValidationError while /health still returns 200. Unauthenticated remote attackers can send one request with an oversized min_tokens to hang all subsequent generation requests until manual restart.","score":7.5,"severity":"high","published":"2026-10-21T00:36:28.373Z","modified":"2026-10-28T15:15:55.536Z","withdrawn":false,"knownExploited":false,"affected":[],"products":[{"vendor":"vllm-project","name":"vllm","platforms":[],"versions":[{"version":"0","status":"affected","lessThan":"0.31.0","versionType":"semver"},{"version":"0.31.0","status":"unaffected","versionType":"semver"}],"defaultStatus":"unaffected","packageURL":"pkg:pypi/vllm","packageIdentity":{"ecosystem":"PyPI","name":"vllm"},"cpes":[],"source":"VulnCheck","url":"https://www.cve.org/CVERecord?id=CVE-2026-114051"}],"remediation":[],"references":["https://github.com/vllm-project/vllm/security/advisories/GHSA-4xqp-c3mv-qff7","https://www.vulncheck.com/advisories/vllm-before-0.31.0-denial-of-service-via-min-tokens-in-inference-v1-generate"],"sources":[{"name":"CVE","id":"CVE-2026-114051","modified":"2026-10-28T15:15:55.536Z","url":"https://www.cve.org/CVERecord?id=CVE-2026-114051"}],"assessments":[{"source":"VulnCheck","vector":"CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N","score":7.1},{"source":"VulnCheck","vector":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H","score":7.5}],"sourceAffected":[],"observedAt":"2026-10-29T14:33:28.480Z","kev":{"status":"not-listed","checkedAt":"2026-10-31T10:20:31.211Z"},"coverage":[{"source":"NVD","status":"not-checked"},{"source":"GHSA","status":"not-checked"},{"source":"EUVD","status":"not-checked"},{"source":"EPSS","status":"not-checked"}],"evidence":[],"aiSummary":{"text":"vLLM before 0.31.0 is vulnerable to a denial of service via min_tokens in /inference/v1/generate, allowing unauthenticated remote attackers to hang generation requests. The issue occurs when a null max_tokens is provided, skipping the min_tokens check. This causes the EngineCore input thread to die on an uncaught VLLMValidationError.","model":"@cf/meta/llama-3.3-70b-instruct-fp8-fast","generatedAt":"2026-10-30T09:13:53.479Z","sources":[{"name":"CVE","url":"https://www.cve.org/CVERecord?id=CVE-2026-114051"}]}}