{
  "cveId": "CVE-2025-49847",
  "url": "https://patchblox.com/security/cve/CVE-2025-49847",
  "summary": "CVE-2025-49847 is fixed in llama.cpp on 1 platform by DEBIAN-CVE-2025-49847 (version 5713\u002Bdfsg-1), released June 17, 2025. It is not reported as exploited. Patch Radar priority: High.",
  "title": "llama.cpp: llama.cpp is an inference of several LLM models in C/C\u002B\u002B",
  "description": "llama.cpp is an inference of several LLM models in C/C\u002B\u002B. Prior to version b5662, an attacker\u2010supplied GGUF model vocabulary can trigger a buffer overflow in llama.cpp\u2019s vocabulary\u2010loading code. Specifically, the helper _try_copy in llama.cpp/src/vocab.cpp: llama_vocab::impl::token_to_piece() casts a very large size_t token length into an int32_t, causing the length check (if (length \u003C (int32_t)size)) to be bypassed. As a result, memcpy is still called with that oversized size, letting a malicious model overwrite memory beyond the intended buffer. This can lead to arbitrary memory corruption and potential code execution. This issue has been patched in version b5662.",
  "published": "2025-06-17T20:15:32Z",
  "severity": "high",
  "cvss": 8.8,
  "cvssVector": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H",
  "epss": 0.00515,
  "epssPercentile": 0.44127,
  "priority": "High",
  "priorityTier": 3,
  "exploited": false,
  "vendorExploited": false,
  "disclosed": false,
  "fixes": [
    {
      "platform": "linux",
      "product": "llama.cpp",
      "release": "Debian 14",
      "fixedVersion": "5713\u002Bdfsg-1",
      "fix": "DEBIAN-CVE-2025-49847",
      "url": "https://patchblox.com/security/fix/linux/debian-cve-2025-49847",
      "released": "2025-06-17T20:15:32Z"
    }
  ],
  "noFixYet": [],
  "dataAsOf": "2026-11-08T19:12:23Z",
  "attribution": {
    "text": "Data: Patchblox Patch Radar",
    "url": "https://patchblox.com/security/patch-radar",
    "terms": "https://patchblox.com/security/patch-radar/use-the-data"
  }
}