{"id":"DEBIAN-CVE-2025-49847","details":"llama.cpp is an inference of several LLM models in C/C++. Prior to version b5662, an attacker‐supplied GGUF model vocabulary can trigger a buffer overflow in llama.cpp’s vocabulary‐loading 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.","modified":"2026-08-27T23:05:33.527802775Z","published":"2025-06-17T20:15:32.437Z","upstream":["CVE-2025-49847"],"references":[{"type":"ADVISORY","url":"https://security-tracker.debian.org/tracker/CVE-2025-49847"}],"affected":[{"package":{"name":"llama.cpp","ecosystem":"Debian:14","purl":"pkg:deb/debian/llama.cpp?arch=source&distro=forky"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"5713+dfsg-1"}]}],"versions":["5151+dfsg-1~exp2","5151+dfsg-1~exp3","5318+dfsg-1","5318+dfsg-2"],"ecosystem_specific":{"urgency":"not yet assigned"},"database_specific":{"source":"https://storage.googleapis.com/osv-test-debian-osv/debian-cve-osv/DEBIAN-CVE-2025-49847.json"}}],"schema_version":"1.9.0","severity":[{"type":"CVSS_V3","score":"CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H"}]}