vLLM: Out-of-vocabulary prompt token id passes input validation and triggers a CUDA device-side assert (engine crash / DoS)

{
  "id": "4361e2ae712f895394e4799feb2fc030e773541b68719c2da0ef1ed25b01093b",
  "url": "https://www.cve.org/CVERecord?id=CVE-2026-115916",
  "method": "structured",
  "payload": {
    "kind": "advisory",
    "title": "vLLM: Out-of-vocabulary prompt token id passes input validation and triggers a CUDA device-side assert (engine crash / DoS)",
    "summary": "vLLM is an inference and serving engine for large language models. Prior to 0.32.0, vllm/tokenizers/hf.py computes max_token_id with an off-by-one upper bound, allowing a pre-tokenized prompt ID equal to the vocabulary size to pass validation. For a tensor-parallel size of one and an exact multiple-of-64 vocabulary, VocabParallelEmbedding.forward does not mask that value before F.embedding uses it, causing a CUDA device-side assertion and terminating the engine. This issue is fixed in version 0.32.0.",
    "evidence": "cve:CVE-2026-115916",
    "products": [
      {
        "name": {
          "value": "vllm",
          "evidence": "/containers/cna/affected/0/product"
        },
        "prerequisites": [],
        "fixed_versions": [],
        "affected_versions": []
      }
    ],
    "severity": "medium",
    "structured": {
      "id": "CVE-2026-115916",
      "aliases": [
        "CVE-2026-115916"
      ],
      "metrics": [
        {
          "cvssV3_1": {
            "scope": "UNCHANGED",
            "version": "3.1",
            "baseScore": 6.5,
            "attackVector": "NETWORK",
            "baseSeverity": "MEDIUM",
            "vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H",
            "integrityImpact": "NONE",
            "userInteraction": "NONE",
            "attackComplexity": "LOW",
            "availabilityImpact": "HIGH",
            "privilegesRequired": "LOW",
            "confidentialityImpact": "NONE"
          }
        }
      ],
      "affected": [
        {
          "vendor": "vllm-project",
          "product": "vllm",
          "versions": [
            {
              "status": "affected",
              "version": "< 0.32.0"
            }
          ]
        }
      ],
      "modified": "2026-10-26T17:18:17.582Z",
      "provider": "cve",
      "references": [
        "https://github.com/vllm-project/vllm/security/advisories/GHSA-q97c-p368-75c4",
        "https://github.com/vllm-project/vllm/pull/59491",
        "https://github.com/vllm-project/vllm/commit/37cecc8141e2dcba9dd534897a164d4baaf27229",
        "https://github.com/vllm-project/vllm/releases/tag/v0.32.0"
      ],
      "provider_metadata": {
        "cna": {
          "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
          "shortName": "GitHub_M",
          "dateUpdated": "2026-10-26T17:18:17.582Z"
        }
      }
    },
    "related_cves": [
      {
        "value": "CVE-2026-115916",
        "evidence": "/cveMetadata/cveId"
      }
    ],
    "security_relevant": true,
    "severity_evidence": "/containers/cna/metrics"
  },
  "source_id": "core-cve",
  "document_id": 322610,
  "published_at": "2026-10-26T17:18:17.582+00:00"
}