Tracked topic
… It includes ML inference frameworks like llama.cpp's Vulkan backend, ncnn, and MNN, which can now use the GPU for compute on top of the CPU though I had to build the newest Vulkan headers to get that working . …
… So, I installed the Vulkan drivers and Cmake tools with apt install glslc glslang-tools libvulkan1 vulkan-tools libvulkan-dev spirv-tools build-essential git cmake curl . …
… Both correctly tracked down draft PR 22673 in the llama.cpp repo, flagged that native MTP isn't merged to master yet, and both correctly described the Vulkan situation: garbage output in early testing on a Radeon RX 9700, the author later updating the comment to say Vulkan and Metal had been tested. …
… For Direct3D games, DXVK or VKD3D-Proton still converts those calls to Vulkan first. …
… Direct3D calls have to go somewhere, and on Linux, the most practical modern target is Vulkan. DXVK translates those Direct3D calls to Vulkan, which Linux handles natively and efficiently. …
… Ollama does have experimental Vulkan support now, which can technically work with Arc GPUs on Windows and Linux, but it's not the polished experience you'd get with CUDA or even ROCm, and as we've shown, performance is significantly worse when comparing XPU with Vulkan. …
… Proton was responsible for translating Windows games to Linux, but Vulkan was what made it possible without running into terrible performance. …
… Here, translating heavy DX12 workloads into Vulkan can introduce higher CPU overhead, resulting in significant performance drops compared to native, day-one Windows execution. …
… Imagination's proprietary DDK already does Vulkan 1.3 on this hardware today, but it doesn't ship in any distro by default. The upstream Mesa PowerVR driver advertises Vulkan 1.2 non-conformant , with Khronos 1.2 conformance submissions due in early 2026. …
… A OpenGL B Vulkan C DirectX D Glide Correct! …