{"data":{"id":"11bfbfcd-b80a-4718-8886-4c192f03d9a2","title":"CVE-2026-115916: vLLM is an inference and serving engine for large language models. Prior to 0.32.0, vllm/tokenizers/hf.py computes…","summary":"vLLM versions prior to 0.32.0 contain an off-by-one error in vllm/tokenizers/hf.py, where max_token_id is computed with an incorrect upper bound. A pre-tokenized prompt ID equal to the vocabulary size passes validation, and with tensor-parallel size one and a vocabulary that is an exact multiple of 64, VocabParallelEmbedding.forward does not mask it before F.embedding uses it. This triggers a CUDA device-side assertion that terminates the engine.","solution":"Fixed in 0.32.0.","labels":["security"],"sourceUrl":"https://nvd.nist.gov/vuln/detail/CVE-2026-115916","publishedAt":"2026-10-26T18:17:39.080Z","cveId":"CVE-2026-115916","cweIds":["CWE-129","CWE-193"],"cvssScore":"6.5","cvssSeverity":"medium","severity":"medium","attackType":["denial_of_service"],"issueType":"vulnerability","affectedPackages":null,"affectedPackageNames":null,"affectedPackageRefs":null,"affectedVendors":[],"affectedVendorsRaw":["vLLM"],"classifierModel":"claude-haiku-5-5","classifierPromptVersion":"v4","summaryPromptVersion":"v2","headline":"vLLM crash from off-by-one token ID validation in pre-tokenized prompts","headlinePromptVersion":"h1","cvssVector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H","attackVector":"network","attackComplexity":"low","privilegesRequired":"low","userInteraction":"none","exploitMaturity":"unknown","epssScore":0,"epssCheckedAt":"2026-10-27T06:10:23.080Z","kevDateAdded":null,"advisoryAliases":null,"affectedPackagesSource":null,"affectedPackagesCheckedAt":"2026-10-27T06:10:29.767Z","patchAvailable":null,"disclosureDate":"2026-10-26T18:17:39.080Z","capecIds":null,"crossRefCount":0,"attackSophistication":"trivial","impactType":["availability"],"aiComponentTargeted":"inference","llmSpecific":true,"classifierConfidence":0.95,"researchCategory":null,"atlasIds":null}}