HIGHđŸ‡”đŸ‡± Wersja polska

CVE-2025-49847

CVSS 8.8v3.1pub. 2025-06-17upd. 2025-08-27

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.

CVSS Vector
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H
  • Ggml Llama.cpp

    APP
    Ggml
    < b5662
đŸ””
CHECK WITH VENDOR
No clear patch data available. Check vendor references.
Tags
RCEMemory
CWE
References

Related vulnerabilities

CVE-2026-34159CRITICAL9.8PL ✓ten sam produkt

RCE w llama.cpp — brak walidacji granic w RPC backend (deserialization)

CVE-2024-42479CRITICAL10.0PL ✓ten sam produkt

Zapis pod dowolny adres pamięci w llama.cpp poprzez wskaĆșnik rpc_tensor

CVE-2026-33298HIGH7.8ten sam produkt

llama.cpp is an inference of several LLM models in C/C++. Prior to b7824, an integer overflow vulnerability in...

CVE-2026-27940HIGH7.8ten sam produkt

llama.cpp is an inference of several LLM models in C/C++. Prior to b8146, the gguf_init_from_file_impl() in gg...

CVE-2026-21869HIGH8.8ten sam produkt

llama.cpp is an inference of several LLM models in C/C++. In commits 55d4206c8 and prior, the n_discard parame...

CVE-2025-49847 — Ggml — Llama.Cpp — HIGH — CVSS 8.8 | CVEbaza.pl