attention

sglang

+740

SGLang is a high-performance serving framework for large language models and multimodal models.

flashinfer

+71

FlashInfer: Kernel Library for LLM Serving

annotated_deep_learning_paper_implementations

+15

🧑‍🏫 60+ Implementations/tutorials of deep learning papers with side-by-side notes 📝; including transformers (original, xl, switch, feedback, vit, ...), optimizers (adam, adabelief, sophia, ...), gans(cyclegan, stylegan2, ...), 🎮 reinforcement learning (ppo, dqn), capsnet, distillation, ... 🧠

tiny-vllm

+11

Build your own high performance LLM inference engine in C++ and CUDA - a smaller version of vLLM

open-jev

+6

an open-source, from-first-principles reconstruction of the ideas behind TypeSafe AI's Jev, written in pytorch

ComfyUI-sol-attn

+4

NVIDIA Sol-Attn for ComfyUI / Triton kernel on SM89 - SM121, with zero-copy MiniMax H3 nodes: memory-efficient attention, scheduled tau with graph preview, and feed-forward chunking. Measured 1.14–1.44× vs SageAttention and −37% MLP peak VRAM on H3

awesome-multi-agent-papers

+2

A compilation of the best multi-agent papers

nlp_paper_study

+1

该仓库主要记录 NLP 算法工程师相关的顶会论文研读笔记