CVE-2024-42479Published Aug 12, 2024 · Aug 12Updated Oct 28
llama.cpp allows write-what-where in rpc_server::set_tensor
ggerganov llama.cpp
Critical10.0Fix available
Affected versions
| Product | Affected | Fixed in |
|---|---|---|
| ggerganov llama.cpp |
| No fix yet |
Original description
llama.cpp provides LLM inference in C/C++. The unsafe `data` pointer member in the `rpc_tensor` structure can cause arbitrary address writing. This vulnerability is fixed in b3561.
From the CVE record · GitHub_M