A modern GUI client based on Tauri, designed to run in Windows, macOS and Linux for tailored proxy experience

A fast communication-overlapping library for tensor/expert parallelism on GPUs.

Distributed Compiler based on Triton for Parallel Systems

My learning notes/codes for ML SYS.

Sampling profiler for Python programs

Trinity-RFT is a general-purpose, flexible and scalable framework designed for reinforcement fine-tuning (RFT) of large language models (LLM).

A fast, clean, responsive Hugo theme.

The Minimum Viable Model website and Jekyll theme.

Efficient Triton Kernels for LLM Training

A compiler for pl0 , c++

The official repo of MiniMax-Text-01 and MiniMax-VL-01, large-language-model & vision-language-model based on Linear Attention

基于Python的开源量化交易平台开发框架

A simple tool to update bib entries with their official information (e.g., DBLP or the ACL anthology).

NCCL Tests

HLLM: Enhancing Sequential Recommendations via Hierarchical Large Language Models for Item and User Modeling

Best practice for training LLaMA models in Megatron-LM

Tensors and Dynamic neural networks in Python with strong GPU acceleration

毒奶自用，懒人配置文件（Quantumult X）：去广告分流规则、Tiktok解锁重写、VSCO解锁、神机分流、blackmatrix7分流规则。

毒奶去网页广告计划用户脚本 For Quantumult X & Surge & Shadowrocket & Loon & Stash & 油猴 ；1.新增页面右下角导航；2.通过调用 JavaScript 移除特定网站网页广告 —— 搜索引擎（Bing/Google）广告及内容农场结果清除/低端影视/欧乐影院/iyf爱壹帆/哔滴影视/Pornhub/Javbus/Supjav/Jable…

Ongoing research training transformer models at scale

A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on Hopper, Ada and Blackwell GPUs, to provide better performance…

BLAS-like Library Instantiation Software Framework

Ongoing research training transformer language models at scale, including: BERT & GPT-2

Zero Bubble Pipeline Parallelism

Development repository for the Triton language and compiler

Enabling PyTorch on XLA Devices (e.g. Google TPU)

Main repo to keep scripts, dockerfiles, wiki, etc