Steven

ILLIXR loads the plugins by the order they are passed in right now. Dependency conflicts easily arise for plugins requesting other services and better handling is required. For now, the...

For long, gldemo and vkdemo have served as testing applications that are not meant to be used on a real headset. However, with the coming support of direct mode, we...

Distributing tensors across NUMA nodes

3

I'm wondering how much support Neural Speed has for NUMA systems. The Advanced Usage page suggests that all tensors should be allocated on the first NUMA node `numactl -m 0...

Running Q4_K_M gguf models: unrecognized tensor type 12

1

``` Welcome to use the llama on the ITREX! AVX:1 AVX2:1 AVX512F:0 AVX_VNNI:1 AVX512_VNNI:0 AMX_INT8:0 AMX_BF16:0 AVX512_BF16:0 AVX512_FP16:0 Loading the bin file with GGUF format... main: seed = 1712361979 model.cpp:...