byungsoo-oh

byungsoo-oh's Stars

• Dao-AILab/flash-attention

  Fast and memory-efficient exact attention

  Language:Python15k1231.2k1.4k

• microsoft/LoRA

  Code for loralib, an implementation of "LoRA: Low-Rank Adaptation of Large Language Models"

  Language:Python11.1k70108698

• vosen/ZLUDA

  CUDA on non-NVIDIA GPUs

  Language:Rust10.3k141183671

• SJTU-IPADS/PowerInfer

  High-speed Large Language Model Serving on PCs with Consumer-grade GPUs

  Language:C++8k78172418

• sgl-project/sglang

  SGLang is a fast serving framework for large language models and vision language models.

  Language:Python7.2k64873690

• NVIDIA/cutlass

  CUDA Templates for Linear Algebra Subroutines

  Language:C++6k1111.2k1k

• mosaicml/composer

  Supercharge Your Model Training

  Language:Python5.2k49552429

• mosaicml/llm-foundry

  LLM training code for Databricks foundation models

  Language:Python4.1k48387536

• cybertronai/gradient-checkpointing

  Make huge neural nets fit in memory

  Language:Python2.7k8242272

• google-deepmind/gemma

  Open weights LLM from Google DeepMind.

  Language:Python2.6k3335328

• databricks/dbrx

  Code examples and resources for DBRX, a large language model developed by Databricks

  Language:Python2.5k4224239

• HazyResearch/ThunderKittens

  Tile primitives for speedy kernels

  Language:Cuda1.9k343493

• NUS-HPC-AI-Lab/OpenDiT

  OpenDiT: An Easy, Fast and Memory-Efficient System for DiT Training and Inference

  Language:Python1.4k236093

• RahulSChand/gpu_poor

  Calculate token/s & GPU memory requirement for any LLM. Supports llama.cpp/ggml/bnb/QLoRA quantization

  Language:JavaScript1.2k71662

• mosaicml/streaming

  A Data Streaming Library for Efficient Neural Network Training

  Language:Python1.2k25185148

• myshell-ai/JetMoE

  Reaching LLaMA2 Performance with 0.1M Dollars

  Language:Python96681080

• pjlab-sys4nlp/llama-moe

  ⛷️ LLaMA-MoE: Building Mixture-of-Experts from LLaMA with Continual Pre-training (EMNLP 2024)

  Language:Python90882249

• volcengine/veScale

  A PyTorch Native LLM Training Framework

  Language:Python690331836

• alibaba/Megatron-LLaMA

  Best practice for training LLaMA models in Megatron-LM

  Language:Python63876455

• forhaoliu/ringattention

  Transformers with Arbitrarily Large Context

  Language:Python62561648

• LLMServe/DistServe

  Disaggregated serving system for Large Language Models (LLMs).

  Language:Jupyter Notebook43554550

• sail-sg/zero-bubble-pipeline-parallelism

  Zero Bubble Pipeline Parallelism

  Language:Python30772916

• microsoft/mscclpp

  MSCCL++: A GPU-driven communication stack for scalable AI applications

  Language:C++2811910543

• UpstageAI/evalverse

  The Universe of Evaluation. All about the evaluation for LLMs.

  Language:Python22181025

• efeslab/fiddler

  Fast Inference of MoE Models with CPU-GPU Orchestration

  Language:Python1788216

• microsoft/ParrotServe

  [OSDI'24] Serving LLM-based Applications Efficiently with Semantic Variable

  Language:Python130549

• TorchMoE/MoE-Infinity

  PyTorch library for cost-effective, fast and easy serving of MoE models.

  Language:Python1127189

• parasailteam/coconet

  Language:HTML723911

• mental2008/awesome-papers

  Here are my personal paper reading notes (including cloud computing, resource management, systems, machine learning, deep learning, and other interesting stuffs).

  608833

• raymin0223/fast_robust_early_exit

  Fast and Robust Early-Exiting Framework for Autoregressive Language Models with Synchronized Parallel Decoding (EMNLP 2023 Long)

  Language:Python56298