ggerganov llama.cpp

• CVE-2024-42479 Critical 10.0 llama.cpp allows write-what-where in rpc_server::set_tensor 2.6%
• CVE-2024-42478 Medium 5.3 llama.cpp allows Arbitrary Address Read in rpc_server::get_tensor 0.6%
• CVE-2024-42477 Medium 5.3 llama.cpp global-buffer-overflow in ggml_type_size 0.5%
• CVE-2024-41130 Medium 5.4 llama.cpp null pointer dereference in gguf_init_from_file 0.3%
• CVE-2024-32878 High 7.1 Use of Uninitialized Variable Vulnerability in llama.cpp 0.7%