Someone should tell Qualcomm then: https://www.qualcomm.com/developer/blog/2024/11/introducing-new-opn-cl-gpu-backend-llama-cpp-for-qualcomm-adreno-gpu https://

Someone should tell Qualcomm then: https://www.qualcomm.com/developer/blog/2024/11/introducing-new-opn-cl-gpu-backend-llama-cpp-for-qualcomm-adreno-gpu https://www.qualcomm.com/developer/blog/2024/11/introducing-... I believe the llama.cpp Vulkan backend is inoperable on Adreno GPUs