CVE-2025-49847

Severity CVSS v4.0:
Pending analysis
Type:
CWE-119 Buffer Errors
Publication date:
17/06/2025
Last modified:
17/06/2025

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