中文

// CVE · PUBLISHED 2026-08-06

CVE-2026-43629

ggml-org / llama.cpp

9.2 · CVSS 4.0Critical

Description

llama.cpp builds b4882 through b9058 contain a heap buffer overflow vulnerability in the KV cache state restore path where the state_read_data() function computes write size without overflow checking, allowing attackers with write access to the slot_save_path directory to corrupt heap memory. Attackers can craft malicious state files where cell_count multiplication overflows or exceeds tensor buffer allocation to write attacker-controlled bytes past buffer boundaries, potentially resulting in heap metadata corruption, model weight corruption, or arbitrary code execution via function pointer overwrite.

References

Related CVEs

← CVE Index: search every CVE from the last 30 days

This product uses the NVD API but is not endorsed or certified by the NVD. CVSS scores do not establish exploitation.