CVE-2025-49847
llama.cpp Vulnerable to Buffer Overflow via Malicious GGUF Model
HIGH 8.8
At a glance
- Source
- NVD
- EPSS
- EPSS 0.52 % · percentile 44 ()
- State
- PUBLISHED
- Published
- Updated
- Assigner
- GitHub_M
- NVD status
- Analyzed
Description
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 < (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.
Weaknesses (CWE)
References
Scores
| Version | Score | Vector | Scored by |
|---|---|---|---|
| CVSS 3.1 | HIGH 8.8 | CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H | GitHub_M |
| CVSS 3.1 | HIGH 8.8 | CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H | NVD · security-advisories@github.com (Secondary) |
European references
- ENISA EUVD · EUVD-2025-18632
- GCVE · GCVE-0-2025-49847
Affected configurations (NVD CPE)
cpe:2.3:a:ggml:llama.cpp:*:*:*:*:*:*:*:*< b5662
Affected products
| Product | Versions |
|---|---|
| ggml-org · llama.cpp | affected < b5662 |
Sources
| Source | Licence |
|---|---|
| CVE List (CVEProject cvelistV5) | CVE Program Terms of Use |
| EPSS (FIRST.org) | FIRST — attribution requested |
| ENISA EUVD | ENISA — reproduction authorised, source acknowledged |
| NVD (NIST) via fkie-cad/nvd-json-data-feeds | Public domain — this product uses data from the NVD API but is not endorsed or certified by the NVD |