Open deep learning compiler stack for cpu, gpu and specialized accelerators

Yet another re-implement of jetson-containers, targeting for Jetson Thor, Spark, and x86.

FB (Facebook) + GEMM (General Matrix-Matrix Multiplication) - https://code.fb.com/ml-applications...

Machine Learning Containers for NVIDIA Jetson and JetPack-L4T

A high-throughput and memory-efficient inference and serving engine for LLMs

An Industrial-Level Controllable and Efficient Zero-Shot Text-To-Speech System

DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling

Fast and memory-efficient exact attention