Pinned Loading

1. Tensile Tensile Public

   Forked from ROCm/Tensile

   Stretching GPU performance for GEMMs and tensor contractions.

   Python

2. fp8_quant_simulation fp8_quant_simulation Public

   analytical model for fp8 quantization (power of exponent)

   Jupyter Notebook

3. LLM-Inference-Modeling LLM-Inference-Modeling Public

   throughput inference modeling for AMD GPU(s)

   Jupyter Notebook

4. LLM_scaling_laws LLM_scaling_laws Public

   code for published papers

   Jupyter Notebook

5. composable_kernel composable_kernel Public

   Forked from ROCm/composable_kernel

   Composable C++ Template abstractions for implementing Tensor contraction operators (GEMM, iGEMM).

   C++