TinyTinyTPU: 2×2 systolic-array TPU-style matrix-multiply unit deployed on FPGA(github.com)115 points by Xenograph 9 months ago | 49 comments
OpenTPU: Open-Source Reimplementation of Google Tensor Processing Unit (TPU)(github.com)142 points by walterbell 1 year ago | 21 comments
Show HN: Tiny-vLLM – high performance LLM inference engine in C++ and CUDA(github.com)170 points by yu3zhou4 5 months ago | 15 comments
Show HN: Tiny Diffusion – A character-level text diffusion model from scratch(github.com)137 points by nathan-barry 11 months ago | 17 comments
Show HN: less than 650 LOC trainable GPT only using NumPy(github.com)77 points by joennlae 2 years ago | 15 comments