{"data":{"id":"7b1c4c71-e190-4cd0-903e-cacb55e34fae","title":"CVE-2026-94626: vLLM through 0.29.0 fails to validate the tp_size parameter in kv_transfer_params on OpenAI-compatible completion endpoi","summary":"vLLM (a popular open-source LLM serving framework) versions up to 0.29.0 don't properly check the tp_size parameter (a setting that controls how a model is split across multiple processors) when handling requests through its OpenAI-compatible API endpoints. This allows attackers to submit extremely large tp_size values that cause the system to use unlimited memory, crashing the decode worker (the part that generates responses) through an out-of-memory error.","solution":"N/A -- no mitigation discussed in source.","labels":["security"],"sourceUrl":"https://nvd.nist.gov/vuln/detail/CVE-2026-94626","publishedAt":"2026-09-21T22:17:01.587Z","cveId":"CVE-2026-94626","cweIds":["CWE-789"],"cvssScore":"7.5","cvssSeverity":"high","severity":"high","attackType":["denial_of_service"],"issueType":"vulnerability","affectedPackages":null,"affectedVendors":[],"affectedVendorsRaw":["vLLM"],"classifierModel":"claude-haiku-4-5-20251001","classifierPromptVersion":"v3","cvssVector":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H","attackVector":"network","attackComplexity":"low","privilegesRequired":"none","userInteraction":"none","exploitMaturity":"unknown","epssScore":0,"patchAvailable":null,"disclosureDate":"2026-09-21T22:17:01.587Z","capecIds":null,"crossRefCount":0,"attackSophistication":"trivial","impactType":["availability"],"aiComponentTargeted":"inference","llmSpecific":true,"classifierConfidence":0.95,"researchCategory":null,"atlasIds":null}}