ggml‑org released llama.cpp version b12340. The update patches a server‑side request forgery (SSRF) issue in its CORS proxy and media URLs. It also ships pre‑built binaries for macOS Apple Silicon, macOS Intel, iOS, and several Ubuntu CPU and Vulkan configurations. The files are available for download from the GitHub release page.
Developers can now run the updated runtime without the previously reported SSRF vulnerability and have ready‑to‑use binaries for a wider range of platforms.
Why it matters
Developers can safely deploy llama.cpp servers and more easily run the model on macOS, iOS and Ubuntu systems.
