{
  "id": "cve-2026-105753",
  "title": "Mirrored multimodal cache desynchronization can crash the shared vLLM EngineCore",
  "product": "vLLM",
  "ai_system": "GPT-5.5-Cyber",
  "organization": "Patch the Planet (Trail of Bits + OpenAI collaboration) / vLLM maintainers",
  "summary": "The default multimodal sender cache records image metadata before a prompt-length admission check, but a rejected request can prevent the engine receiver cache from receiving the payload. A later request that reuses the image hash is sent as a cache-only reference, and the receiver assertion can terminate the shared EngineCore.",
  "ai_role": "The primary vLLM advisory identifies Patch the Planet (Trail of Bits + OpenAI collaboration) and states this vulnerability was discovered using GPT-5.5-Cyber. The advisory and linked PR describe this exact cache desynchronization; no autonomous or independent reproduction claim is made.",
  "disclosure_date": "2026-09-28",
  "ai_attribution_date": "2026-09-28",
  "classification": "disclosure",
  "classification_note": "The vLLM security advisory provides a public disclosure date and describes the vulnerability. It does not report exploitation in the wild or establish whether the issue was unknown to defenders before discovery.",
  "cves": [
    "CVE-2026-105753"
  ],
  "source_url": "https://github.com/vllm-project/vllm/security/advisories/GHSA-ph3r-5jfg-f84f",
  "ai_source_url": "https://github.com/vllm-project/vllm/security/advisories/GHSA-ph3r-5jfg-f84f",
  "advisory_url": "https://github.com/vllm-project/vllm/security/advisories/GHSA-ph3r-5jfg-f84f",
  "remediation": "fix documented",
  "remediation_note": "The primary vLLM advisory lists 0.28 as the patched version boundary. It links a proposed fix; this record does not imply that any deployed instance has installed an update.",
  "checked_date": "2026-11-17",
  "cvss": [
    {
      "cve": "CVE-2026-105753",
      "version": "3.1",
      "score": 6.5,
      "vector": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H",
      "source_url": "https://github.com/vllm-project/vllm/security/advisories/GHSA-ph3r-5jfg-f84f",
      "source_name": "vllm-project/vllm GHSA-ph3r-5jfg-f84f",
      "checked_date": "2026-11-17"
    }
  ],
  "impacts": [
    {
      "cves": [
        "CVE-2026-105753"
      ],
      "consequences": [
        "denial of service"
      ],
      "conditions": "An authenticated request against a vLLM deployment using the default mirrored LRU multimodal cache can trigger prompt rejection after rendering; a later request reusing that media hash can crash the shared engine. The advisory excludes processor_only, disabled-caching and shared-memory cache paths.",
      "exploitation": "The advisory describes a remotely reachable request sequence and a shared EngineCore crash; it does not report exploitation in the wild.",
      "source_url": "https://github.com/vllm-project/vllm/security/advisories/GHSA-ph3r-5jfg-f84f",
      "checked_date": "2026-11-17"
    }
  ]
}
