CVE-2025-49847

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.

CVSS v3.1 Score

8.8

HIGH

CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H

Weakness Type (CWE)

CWE-119 Buffer Overflow

CWE-195 CWE-195

Affected Products

Vendor	Product	Versions
ggml	llama.cpp	< b5662

References

Advisories & Patches

https://github.com/ggml-org/llama.cpp/commit/3cfbbdb44e08fd19429fed6cc85b982a91f0efd5 https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-8wwf-w4qm-gpqr

Frequently Asked Questions

What is CVE-2025-49847? +

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. It has a CVSS v3.1 base score of 8.8 (HIGH).

How severe is CVE-2025-49847? +

CVE-2025-49847 has a CVSS v3.1 score of 8.8 out of 10, rated HIGH. This is a high-severity vulnerability that should be prioritized for patching.

What products are affected by CVE-2025-49847? +

CVE-2025-49847 affects products from ggml, specifically: llama.cpp. Check the affected products table above for specific version ranges.

How do I check if I'm vulnerable to CVE-2025-49847? +

You can use Secably's free Website Scanner to check your website for known vulnerabilities. For infrastructure scanning, use the Port Scanner to identify exposed services that may be affected. Check the vendor advisories linked above for specific patch and version information.

Related Vulnerabilities

CVE-2024-11139

CWE-119: Improper Restriction of Operations within the Bounds of a Memory Buffer vulnerability exists that could allow local attackers to …

CVE-2025-1866

Improper Restriction of Operations within the Bounds of a Memory Buffer vulnerability in warmcat libwebsockets allows Pointer Manipulation, potentially leading …

CVE-2025-2401

Buffer overflow vulnerability in Immunity Debugger affecting version 1.85, its exploitation could allow a local attacker to execute arbitrary code, …

CVE-2025-46333

z2d is a pure Zig 2D graphics library. Versions of z2d after `0.5.1` and up to and including `0.6.0`, when …

CVE-2025-6151

A vulnerability has been found in TP-Link TL-WR940N V4 and TL-WR841N V11. Affected by this issue is some unknown functionality …

CVE-2025-34096

A stack-based buffer overflow vulnerability exists in Easy File Sharing HTTP Server version 7.2. The flaw is triggered when a …