Stars
本科华五，曾赴美qs50读博，某兄弟院校副教授，校园门卫亭女性主理人，为防止炸号的备份平台，是本人。
Fast and memory-efficient exact attention
懒人配置vim方案，有详细说明，以及现成的.vimrc配置文件供使用。The lazy person configures the vim scheme, with detailed instructions, and the ready-made `.vimrc` configuration file for use.
SGLang is a high-performance serving framework for large language models and multimodal models.
LieGraph - AI Agent-Driven "Who Is Spy" Game. Multi-agent social deduction game built with LangGraph where autonomous AI agents use LLM reasoning to find the spy through conversation analysis and …
A lightweight, specification-first Go API mock server that generates realistic responses from OpenAPI 3.0 specifications.
An Easy-to-use, Scalable and High-performance Agentic RL Framework based on Ray (PPO & DAPO & REINFORCE++ & TIS & vLLM & Ray & Async RL)
Lightning-Fast RL for LLM Reasoning and Agents. Made Simple & Flexible.
Development repository for the Triton language and compiler
My learning notes for ML SYS.
Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.
verl: Volcano Engine Reinforcement Learning for LLMs
Production-ready platform for agentic workflow development.
A Datacenter Scale Distributed Inference Serving Framework
Archived implementation of BLAS using the SYCL open standard. See oneMath for a replacement.
High Performance Inference Serving Engine for popular LLMs based on GPU
Reference models for Intel(R) Gaudi(R) AI Accelerator
Pytorch reimplementation of the Vision Transformer (An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale)
Code to replicate the key results from Exploring the Limits of Out-of-Distribution Detection (https://arxiv.org/abs/2106.03004) by Stanislav Fort, Jie Ren, Balaji Lakshminarayanan, published at Neu…
Source code for 'Data Parallel C++: Mastering DPC++ for Programming of Heterogeneous Systems using C++ and SYCL' by James Reinders, Ben Ashbaugh, James Brodman, Michael Kinsner, John Pennycook, Xin…