CVE-2025-49847

Katalog CVE

CVSS 8.8 · High EPSS 0,52%

Kerentanan Improper Restriction of Operations within the Bounds of a Memory Buffer pada ggml llama.cpp. Skor CVSS 3.1 8.8 (High) menurut github.com.

Deskripsi (NVD)

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.

Rincian

Terbit di NVD

17 Juni 2025

Terakhir diubah

17 Juni 2026

Status NVD

Analyzed

CVSS 3.1

8.8 (High), sumber github.com
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H · uraikan di kalkulator

Jenis kelemahan (CWE)

CWE-119: Improper Restriction of Operations within the Bounds of a Memory Buffer
CWE-195: Signed to Unsigned Conversion Error

EPSS (30 hari)

0,52% peluang dieksploitasi; lebih tinggi dari 42,2% CVE lain

Produk dan versi terdampak

> Data dari konfigurasi CPE di NVD.

Referensi dan perbaikan

• https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-8wwf-w4qm-gpqr mitigasi · advisory vendor
• https://github.com/ggml-org/llama.cpp/commit/3cfbbdb44e08fd19429fed6cc85b982a91f0efd5 patch