CyberRota Analysis
AI-GeneratedThe use-after-free vulnerability in llama.cpp affects multiple tokenization endpoints, allowing attackers to exploit a race condition that can lead to crashes or potential code execution. This issue arises when the main thread frees the vocabulary resource while HTTP worker threads are still accessing it, particularly when the --sleep-idle-seconds option is configured. Organizations utilizing affected versions of llama.cpp should prioritize patching this vulnerability to mitigate the risk of exploitation.
Public Exploit Signal
A public exploit, PoC, GitHub repository or Metasploit reference was detected for this CVE.
Note: these links are listed for security research and verification purposes only.
Original NVD Description
llama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in llama-server affecting six tokenization endpoints (/tokenize, /detokenize, /infill, /apply-template, /rerank, and /anthropic/count_tokens) that bypass the task queue and access ctx_server.vocab directly on HTTP worker threads. Attackers can exploit a time-of-check-time-of-use race condition where the main thread destroys and frees vocab after the synchronization lock is released but before the handler finishes using it, causing a crash or potential code execution when --sleep-idle-seconds is configured.