goku

从零开始训练原生多模态大模型

Example models using DeepSpeed

百度开源中文词法分析工具: 分词，词性标注，命名实体识别

Build a modern LLM from scratch. Every line commented. Explained like we are five.

Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

LimiX: Unleashing Structured-Data Modeling Capability for Generalist Intelligence https://arxiv.org/abs/2509.03505

Agent-R1: Training Powerful LLM Agents with End-to-End Reinforcement Learning

Modern columnar data format for ML and LLMs implemented in Rust. Convert from parquet in 2 lines of code for 100x faster random access, vector index, and data versioning. Compatible with Pandas, DuckDB, Polars, Pyarrow, and PyTorch with more integrations coming..

Spark-TTS Inference Code

SOTA Open Source TTS

GLM-130B: An Open Bilingual Pre-Trained Model (ICLR 2023)

ChatGLM-6B：开源双语对话语言模型 | An Open Bilingual Dialogue Language Model

Next-generation AI Agent Optimization Platform: Cozeloop addresses challenges in AI agent development by providing full-lifecycle management capabilities from development, debugging, and evaluation to monitoring.

An AI agent development platform with all-in-one visual tools, simplifying agent creation, debugging, and deployment like never before. Coze your way to AI Agent creation.

Text-audio foundation model from Boson AI

The Entropy Mechanism of Reinforcement Learning for Large Language Model Reasoning.

An Industrial-Level Controllable and Efficient Zero-Shot Text-To-Speech System

IndexTTS Fine-tuning notebooks

💎 A list of accessible speech corpora for ASR, TTS, and other Speech Technologies

No description.

Convert PDF to markdown + JSON quickly with high accuracy

LLaMA: Open and Efficient Foundation Language Models

Curate better data for LLMs

Fast Semantic Text Deduplication & Filtering

Python & Command-line tool to gather text and metadata on the Web: Crawling, scraping, extraction, output as CSV, JSON, HTML, MD, TXT, XML

Curated list of datasets and tools for post-training.

Summarize existing representative LLMs text datasets.

The LLM Evaluation Framework

Building LLaMA 4 MoE from Scratch

Official inference repo for FLUX.1 models

No fortress, purely open ground. OpenManus is Coming.

🚀 「大模型」1小时从0训练26M参数的视觉多模态VLM！🌏 Train a 26M-parameter VLM from scratch in just 1 hours!

Implement a ChatGPT-like LLM in PyTorch from scratch, step by step

Building system of typhoon cloud computing platform of tencent, support C/C++/protobuf/thrift/lex/yacc/swig.

A GUI Agent application based on UI-TARS(Vision-Lanuage Model) that allows you to control your computer using natural language.

A curated list of 120+ LLM libraries category wise.

No description.

Download web video and audio

A list of AI autonomous agents

No description.

深度学习经典、新论文逐段精读

A reading list on LLM based Synthetic Data Generation 🔥

Fully open reproduction of DeepSeek-R1

Zotero chat PDF with GPT, ChatGPT, Claude, Gemini, DeepSeek

Awesome-LLM: a curated list of Large Language Model

No description.

No description.

AI-Driven Research Assistant: An advanced multi-agent system for automating complex research processes. Leveraging LangChain, OpenAI GPT, and LangGraph, this tool streamlines hypothesis generation, data analysis, visualization, and report writing. Perfect for researchers and data scientists seeking to enhance their workflow and productivity.

Code I wrote for my AI & LLM workshops

企业级RAG系统从入门到精通

收录了若干读博中所遇到的问题和相关资料

parallel corpus dataset from the mnbvc project

Solution for checking file if contain NSFW content.

Prompt optimization scratch

Codebase for Aria - an Open Multimodal Native MoE

Vision model based PDF chunking.

Code of Pyramidal Flow Matching for Efficient Video Generative Modeling

Image composition toolbox: everything you want to know about image composition or object insertion

Educational framework exploring ergonomic, lightweight multi-agent orchestration. Managed by OpenAI Solution team.

Offical Repo for "Programming Every Example: Lifting Pre-training Data Quality Like Experts at Scale"

Writing FLUX in Triton

No description.

File Parser optimised for LLM Ingestion with no loss 🧠 Parse PDFs, Docx, PPTx in a format that is ideal for LLMs.

Freeing data processing from scripting madness by providing a set of platform-agnostic customizable pipeline processing blocks.

Adala: Autonomous DAta (Labeling) Agent framework

Your AI second brain. Get answers to your questions, whether they be online or in your own notes. Use foundation models or private, local LLMs. Self-host locally or use our cloud instance. Access from Obsidian, Emacs, Desktop app, Web or Whatsapp.

OpenRefine is a free, open source power tool for working with messy data and improving it

Open source libraries and APIs to build custom preprocessing pipelines for labeling, training, or production machine learning pipelines.

Open-Sora: Democratizing Efficient Video Production for All

No description.

👮‍♂️The sensitive word tool for java.(敏感词/违禁词/违法词/脏词。基于 DFA 算法实现的高性能 java 敏感词过滤工具框架。请勿发布涉及政治、广告、营销、翻墙、违反国家法律法规等内容。高性能敏感词检测过滤组件，附带繁体简体互换，支持全角半角互换，汉字转拼音，模糊搜索等功能。)

🚀「Douyin_TikTok_Download_API」是一个开箱即用的高性能异步抖音、快手、TikTok、Bilibili数据爬取工具，支持API调用，在线批量解析及下载。

Inference code for LLaMA models

An open-source tool-augmented conversational language model from Fudan University

No description.

Evaluation suite for LLMs

通用考试题库数据集 选择 填空 简答

No description.

MNBVC数学问答数据清洗

MNBVC General Cleaning Script for the Q&A Dataset of Foreign Ministry Journalists

Extract Chinese/English QA Data from WikiHow pages.

No description.

No description.

No description.

No description.

用于提取github-code-zip文件的内容，并保存为jsonl格式

github仓库下载器

No description.

No description.

No description.

文本去重

本项目旨在对大量文本文件进行快速编码检测和转换以辅助mnbvc语料集项目的数据清洗工作

MNBVC(Massive Never-ending BT Vast Chinese corpus)超大规模中文语料集。对标chatGPT训练的40T数据。MNBVC数据集不但包括主流文化，也包括各个小众文化甚至火星文的数据。MNBVC数据集包括新闻、作文、小说、书籍、杂志、论文、台词、帖子、wiki、古诗、歌词、商品介绍、笑话、糗事、聊天记录等一切形式的纯文本中文数据。

🐙 Guides, papers, lecture, notebooks and resources for prompt engineering

No description.

No description.

Convert Compute And Books Into Instruct-Tuning Datasets

No description.

Agent framework and applications built upon Qwen, featuring Code Interpreter and Chrome browser extension.

Collection of training data management explorations for large language models

👁️ + 💬 + 🎧 = 🤖 Curated list of top foundation and multimodal models! [Paper + Code]

OpenAGI: When LLM Meets Domain Experts

Crawlee—A web scraping and browser automation library for Node.js that helps you build reliable crawlers. Fast.

Crawl a site to generate knowledge files to create your own custom GPT from a URL

Data-Copilot: Bridging Billions of Data and Humans with Autonomous Workflow

Skywork series models are pre-trained on 3.2TB of high-quality multilingual (mainly Chinese and English) and code data. We have open-sourced the model, training data, evaluation data, evaluation methods, etc. 天工系列模型在3.2TB高质量多语言和代码数据上进行预训练。我们开源了模型参数，训练数据，评估数据，评估方法。

供AI训练的中文数据集（持续更新。。。），目前的数据集餐饮行业8000问，百度知道，Alpaca中文数据集，计算机领域数据集，Vicuna数据集，RedPajama数据集，Wikipedia中文词条数据集，网站论坛问答数据集

A privacy-first, self-hosted, fully open source personal knowledge management software, written in typescript and golang.

Ray is a unified framework for scaling AI and Python applications. Ray consists of a core distributed runtime and a toolkit of libraries (Ray AIR) for accelerating ML workloads.

AI Code Translator，Generate Code or Natural Language To Programming Language

chinese NLP corpus of chinese science fiction,chinese science fiction corpus : About 4675 Chinese science fiction novels 大约有4675本科幻小说，中文科幻小说自然语言处理语料库，中文科幻小说文本语料库，中文科幻小说文本数据库，科幻小说语料

chinese NLP corpus of chinese science fiction, chinese science fiction corpus: Archive of the Ark Plan of Ula Science Fiction Website 乌拉科幻小说网方舟计划存档，中文科幻小说自然语言处理语料库，中文科幻小说文本语料库，中文科幻小说文本数据库，科幻小说语料

Make Llama2 use Code Execution, Debug, Save Code, Reuse it, Access to Internet

Gorse open source recommender system engine

精益副业：程序员如何优雅地做副业

:books: Freely available programming books

LongLLaMA is a large language model capable of handling long contexts. It is based on OpenLLaMA and fine-tuned with the Focused Transformer (FoT) method.

Implementation of plug in and play Attention from "LongNet: Scaling Transformers to 1,000,000,000 Tokens"

Awesome Pretrained Chinese NLP Models，高质量中文预训练模型集合

A 13B large language model developed by Baichuan Intelligent Technology

DAMO-ConvAI: The official repository which contains the codebase for Alibaba DAMO Conversational AI.

No description.

The official GitHub page for the survey paper "A Survey of Large Language Models".

《AI 研发提效研究：自己动手训练 LoRA》，包含 Llama （Alpaca LoRA）模型、ChatGLM （ChatGLM Tuning）相关 Lora 的训练。训练内容：用户故事生成、测试代码生成、代码辅助生成、文本转 SQL、文本生成代码……

Revolutionizing Database Interactions with Private LLM Technology

🔥 🔥 🔥 An intelligent and versatile general-purpose SQL client and reporting tool for databases which integrates ChatGPT capabilities.(智能的通用数据库SQL客户端和报表工具)

自然语言转SQL，直接连接数据库查询

聚宝盆(Cornucopia): 基于中文金融知识的LLaMA微调模型；涉及SFT、RLHF、GPU训练部署等

大模型高效量化训练+部署 (Efficient Finetuning of QLoRA LLMs). QLoRA, LLama, bloom, baichuan-7B, GLM

Tuning LLMs with no tears💦, sharing LLM-tools with love❤️.

No description.

A large-scale 7B pretraining language model developed by BaiChuan-Inc.

Simple go utility to download HuggingFace Models and Datasets

The repo of Large Lanuage Model (LLM) for recommendation system.

开源SFT数据集整理,随时补充

人教版小学到大学数学

Code and documentation to train Stanford's Alpaca models, and generate the data.

百亿参数的中英文双语基座大模型

Numbers every LLM developer should know

The complete training code of the open-source Chinese-Llama model, including the full process from pre-training instructing and RLHF.

中文LLaMA&Alpaca大语言模型+本地CPU/GPU训练部署 (Chinese LLaMA & Alpaca LLMs)

Official repo for consistency models.

BELLE: Be Everyone's Large Language model Engine（开源中文对话大模型）

An experimental open-source attempt to make GPT-4 fully autonomous.

Tools to download and cleanup Common Crawl data

Source code for Twitter's Recommendation Algorithm

Examples and guides for using the OpenAI API

大麦网抢票脚本

Awesome List of resources on leading people and being a manager. Geared toward tech, but potentially useful to anyone.

Official implementation for "Multimodal Chain-of-Thought Reasoning in Language Models" (stay tuned and more will be updated)

Making big AI models cheaper, easier, and more scalable

Code for the paper "Language Models are Unsupervised Multitask Learners"

⚡️Blazing-Fast Python Search Engine 🐍

30天自制C++服务器，包含教程和源代码

A library for high-performant graph algorithms.

The 2022 Java Developers Roadmap

Named Entity Recognition

Code for EMNLP 2021 main conference paper "Text AutoAugment: Learning Compositional Augmentation Policy for Text Classification"

DomainWordsDict, Chinese words dict that contains more than 68 domains, which can be used as text classification、knowledge enhance task。涵盖68个领域、共计916万词的专业词典知识库，可用于文本分类、知识增强、领域词汇库扩充等自然语言处理应用。

StellarGraph - Machine Learning on Graphs

CCF ADL 2019 slides for knowledge graph fusion

No description.

Source code for MetaKG: Meta-learning on Knowledge Graph for Cold-start Recommendation

Code for KHGT model, AAAI2021

Code for the paper "Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer"

Graph Neural Network Library for PyTorch

Python package built to ease deep learning on graph, on top of existing DL frameworks.

The implementation of the AutoEncoder-coupled Transformer Network by TensorFlow

A tensorflow implementation of Knowledge-aware Graph Neural Networks with Label Smoothness regularization

Code for ALBEF: a new vision-language pre-training method

No description.

Easy-to-use,Modular and Extendible package of deep-learning based CTR models .

Some basic NER algorithms implemented by PyTorch.

A series of Jupyter notebooks that walk you through the fundamentals of Machine Learning and Deep Learning in python using Scikit-Learn and TensorFlow.

The Easiest Knowledge Distillation Library for Lightweight Deep Learning

Learning embeddings for classification, retrieval and ranking.

Aim — a super-easy way to record, search and compare AI experiments

自然语言处理，知识图谱相关语料。按照Task细分，欢迎PR。

A PyTorch implementation of GraphSAGE. This package contains a PyTorch implementation of GraphSAGE.

链家二手房租房在线数据，存量房交易服务平台数据，详细数据分析教程

Code for Paper: CFKG: Learning Heterogeneous Knowledge Base Embeddings for Explainable Recommendation (KBE4ER)

CCKS 2019 Task 2: Entity Recognition and Linking

Python-based tools for document analysis and OCR

OneFlow is a performance-centered and open-source platform for machine learning.

基于Bert（或Robert）预训练模型微调命名实体识别任务，并提供基于docker的Tensor Serving部署模型方案实现过程。

使用预训练语言模型ALBERT做中文NER

A LITE BERT FOR SELF-SUPERVISED LEARNING OF LANGUAGE REPRESENTATIONS, 海量中文预训练ALBERT模型

Chinese NER(Named Entity Recognition) using BERT(Softmax, CRF, Span)

The implementation of Dual Adversarial Neural Transfer (DATNets) for Sequence Labeling, which is able to tackle sequence labeling tasks such as POS Tagging, NER, Chunking,etc..

Read 'Dice Loss for Data-imbalanced NLP Tasks' this evening and try to implement it

No description.

Multinomial Adversarial Networks for Multi-Domain Text Classification (NAACL 2018)

Code and data of the "Multi-domain adversarial learning" paper, Schoenauer-Sebag et al., accepted at ICLR 2019

multi_task_NLP is a utility toolkit enabling NLP developers to easily train and infer a single model for multiple tasks.

Another Domain Adaptation library, aimed at researchers.

Text preprocessing, representation and visualization from zero to hero.

No description.

ACL 2018: Hybrid semi-Markov CRF for Neural Sequence Labeling (http://aclweb.org/anthology/P18-2038)

Reject complicated operations for incorporating lexicon for Chinese NER.

Multi-Grained Named Entity Recognition (ACL 2019)