DeepEP: an efficient expert-parallel communication library

TradingAgents: Multi-Agents LLM Financial Trading Framework

CUDA Templates and Python DSLs for High-Performance Linear Algebra

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…

verl: Volcano Engine Reinforcement Learning for LLMs

一个基于Python和Flask的Web应用，整合了多维度股票分析能力和人工智能辅助决策功能。系统通过AKShare获取股票数据，结合技术分析、基本面分析和资金面分析，为投资者提供全方位的投资决策支持。

slime is an LLM post-training framework for RL Scaling.

A family of open-sourced Mixture-of-Experts (MoE) Large Language Models

Kronos: A Foundation Model for the Language of Financial Markets

📹 A more flexible framework that can generate videos at any resolution and creates videos from images.

Implementation Code for Omni-Effects

Wan: Open and Advanced Large-Scale Video Generative Models

🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.

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

Instruct-tune LLaMA on consumer hardware

A PyTorch Extension: Tools for easy mixed precision and distributed training in Pytorch

Fast and memory-efficient exact attention

MAGI-1: Autoregressive Video Generation at Scale

Official PyTorch implementation of One-Minute Video Generation with Test-Time Training

NCCL Tests

FlashInfer: Kernel Library for LLM Serving

Ring attention implementation with flash attention

A maintained ctags implementation

A collection of LLM papers, blogs, and projects, with a focus on OpenAI o1 🍓 and reasoning techniques.

Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.

TTT Code Release

TTT Code Release

Implementation of Alphafold 3 from Google Deepmind in Pytorch

Langflow is a powerful tool for building and deploying AI-powered agents and workflows.