CVE-2024-58399 | ggml-org llama.cpp GGUF Tensor Offset llama_model_loader tensor offsets out-of-bounds (b2731)
A vulnerability, which was classified as problematic, has been found in ggml-org llama.cpp. This impacts the function llama_model_loader of the component GGUF Tensor Offset Handler. Performing a manipulation of the argument tensor offsets results in out-of-bounds read.
This vulnerability is cataloged as CVE-2024-58399. The attack must be initiated from a local position. There is no exploit available.
Applying a patch is the recommended action to fix this issue.VulDB Recent EntriesRead More