Pinned Repositories
CS6230-project
allo
Allo: A Programming Model for Composable Accelerator Design
llama.cpp
LLM inference in C/C++
allo-kaixin
Allo: A Programming Model for Composable Accelerator Design
Bitnet-simple-end2end-cpp
Simple C++ implementation of Bitnet Attention without multiplication, and omit complex optimizations for specific CPU architectures.
LLM-accelerator
tensat
Re-implementation of the TASO compiler using equality saturation
torchsparse
TorchSparse: Efficient Training and Inference Framework for Sparse Convolution on GPUs.
hqq
Official implementation of Half-Quadratic Quantization (HQQ)
gpt-fast
Simple and efficient pytorch-native transformer text generation in <1000 LOC of python.
kaizizzzzzz's Repositories
kaizizzzzzz/LLM-accelerator
kaizizzzzzz/Bitnet-simple-end2end-cpp
Simple C++ implementation of Bitnet Attention without multiplication, and omit complex optimizations for specific CPU architectures.
kaizizzzzzz/tensat
Re-implementation of the TASO compiler using equality saturation
kaizizzzzzz/torchsparse
TorchSparse: Efficient Training and Inference Framework for Sparse Convolution on GPUs.
kaizizzzzzz/allo-kaixin
Allo: A Programming Model for Composable Accelerator Design
kaizizzzzzz/BiLLM
(ICML 2024) BiLLM: Pushing the Limit of Post-Training Quantization for LLMs
kaizizzzzzz/egg
egraphs good
kaizizzzzzz/gemma_pytorch
The official PyTorch implementation of Google's Gemma models
kaizizzzzzz/gpt-fast
Simple and efficient pytorch-native transformer text generation in <1000 LOC of python.
kaizizzzzzz/lassonet
Feature selection in neural networks
kaizizzzzzz/LMOps
General technology for enabling AI capabilities w/ LLMs and MLLMs
kaizizzzzzz/LoftQ
kaizizzzzzz/punica
Serving multiple LoRA finetuned LLM as one
kaizizzzzzz/pytorchTutorial
PyTorch Tutorials from my YouTube channel
kaizizzzzzz/Sparse-matrix-operation
kaizizzzzzz/stable-baselines3
PyTorch version of Stable Baselines, reliable implementations of reinforcement learning algorithms.
kaizizzzzzz/taso
kaizizzzzzz/kaizizzzzzz.github.io
AcadHomepage: A Modern and Responsive Academic Personal Homepage
kaizizzzzzz/rapidstream-tapa
RapidStream TAPA compiles task-parallel HLS program into high-frequency FPGA accelerators.
kaizizzzzzz/Serpens-new-rapidstream
Serpens is an HBM FPGA accelerator for SpMV