llama.cpp vulnerabilities

Product

llama.cpp

Vendor

ggml

Kind

Inference server

Securaity has published 13 advisories for llama.cpp: 11 High, 2 Medium.

llama.cpp is an inference server: software that loads AI models and answers requests to them from other programs or users. It is maintained by ggml, the open-source project and organization behind it. It is typically installed and run by IT staff or developers who host AI models themselves instead of using a vendor's cloud service. It runs on a server or computer that the organization itself controls.

RSS feed for llama.cpp