Junhui He's repositories
000
llama.cpp
Port of Facebook's LLaMA model in C/C++
Language:CMIT000
nanoGPT
The simplest, fastest repository for training/finetuning medium-sized GPTs.
Language:PythonMIT000
PowerInfer
High-speed Large Language Model Serving on PCs with Consumer-grade GPUs
Language:C++MIT000
tvm
Open deep learning compiler stack for cpu, gpu and specialized accelerators
Language:PythonApache-2.0000
whisper.cpp
Port of OpenAI's Whisper model in C/C++
Language:CMIT000