LLM training in simple, raw C/CUDA

A massively parallel, optimal functional runtime in Rust

DeepEP: an efficient expert-parallel communication library

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

RTP-LLM: Alibaba's high-performance LLM inference engine for diverse applications.

Efficient GPU kernels for block-sparse matrix multiplication and convolution

Single-file, pure CUDA C implementation for running inference on Qwen3 0.6B GGUF. No Dependencies.

Causal set quantum cosmology