hipudding
Results
63
comments of
hipudding
310P will support soon. May I ask what scenario you are using with llama.cpp? We are analyzing the usage scenarios of llama.cpp with Ascend NPU, and I want to collect...
Update doc.
### GGML_CANN_ASYNC_MODE Enables asynchronous operator submission. Disabled by default. ### GGML_CANN_MEM_POOL Specifies the memory pool management strategy: * vmm: Utilizes a virtual memory manager pool. If hardware support for VMM...