FisherKKK/slime
slime is an LLM post-training framework for RL Scaling.
Repositories
FisherKKK/inspire
个人技术知识库
FisherKKK/flashinfer
FlashInfer: Kernel Library for LLM Serving
FisherKKK/sglang
SGLang is a fast serving framework for large language models and vision language models.
FisherKKK/static-constraint-decoding
Sparse Transition Matrix-Accelerated Trie Index for Constrained Decoding (https://arxiv.org/abs/2602.22647)
FisherKKK/mini-sglang
A compact implementation of SGLang, designed to demystify the complexities of modern LLM serving systems.
FisherKKK/nano-vllmx
Nano vLLM
FisherKKK/verl
verl/HybridFlow: A Flexible and Efficient RL Post-Training Framework
FisherKKK/DeepEP
DeepEP: an efficient expert-parallel communication library
FisherKKK/DeepGEMM
DeepGEMM: clean and efficient BLAS kernel library on GPU
FisherKKK/nano-llama-flash
learning by doing. Implement SOA kernels and RL algo to fully grasp performance limits
FisherKKK/cuvs
cuVS - a library for vector search and clustering on the GPU
FisherKKK/TensorRT-LLM
TensorRT LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and supports state-of-the-art optimizations to perform inference efficiently on NVIDIA GPUs. TensorRT LLM also contains components to create Python and C++ runtimes that orchestrate the inference execution in a performant way.
FisherKKK/FasterTransformer
Transformer related optimization, including BERT, GPT
FisherKKK/Liger-Kernel
Efficient Triton Kernels for LLM Training
FisherKKK/flash-linear-attention
🚀 Efficient implementations for emerging model architectures
FisherKKK/jax
Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more
FisherKKK/tensorflow
An Open Source Machine Learning Framework for Everyone
FisherKKK/pytorch
Tensors and Dynamic neural networks in Python with strong GPU acceleration
FisherKKK/lectures
Material for gpu-mode lectures
FisherKKK/vllm
A high-throughput and memory-efficient inference and serving engine for LLMs
FisherKKK/flash-attention
Fast and memory-efficient exact attention
FisherKKK/backup
FisherKKK/cutlass
CUDA Templates and Python DSLs for High-Performance Linear Algebra
FisherKKK/knowiki
knowledge base
FisherKKK/gif
FisherKKK/EvelDB
FisherKKK/ragflow
RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context layer for LLMs
FisherKKK/dragonfly
A modern replacement for Redis and Memcached
FisherKKK/folly
An open-source C++ library developed and used at Facebook.