Explore index

Vulkan Backend

A compute acceleration layer based on the cross-platform Vulkan API, enabling llama.cpp to perform tensor operations on non-NVID

No public content is connected to this entity yet.

Vulkan Backend | Today for AI