Project directory

Projects

Open-source AI projects tracked by the leaderboard, ordered by GitHub stars.

Python 3,142 stars

nidhaloff/igel

a delightful machine learning tool that allows you to train, test, and use models without writing code

Overview: Igel is a user-friendly machine learning tool designed for non-programmers, enabling them to train, test, and utilize models without having to write any code. Key features: - Simplified model training and testing - Automatic model optimization - Integration with various data sources - Accessible documentation and tutorials Tech stack: Python Why it matters: Igel simplifies the complex world of machine learning, making it accessible to everyone, including those without a background in programming. This tool enhances the democratization of AI, making it easier for businesses and individuals to harness the power of machine learning without the need for extensive technical knowledge.

Python 3,140 stars

DAMO-NLP-SG/Video-LLaMA

[EMNLP 2023 Demo] Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding

Overview: Video-LLaMA is an instruction-tuned audio-visual language model designed to enhance the understanding and generation of videos and audio in large language models. Key features: - Supports video and audio understanding and generation. - Uses LLaMA-2 as the language decoder for improved performance. - Available in two versions: Video-LLaMA-2 (LLaMA-2-Chat) and Video-LLaMA-BiLLA (fine-tuned with BiLLa-7B-SFT). - Comes with pre-trained checkpoints and easy-to-use codebase. Tech stack: Python Why it matters: Video-LLaMA leverages advanced language models to improve the capabilities of large language models, enabling them to better understand and interact with videos and audio, making them more versatile and useful in various applications.

Python 3,138 stars

FudanNLP/fastNLP

fastNLP: A Modularized and Extensible NLP Framework. Currently still in incubation.

Overview: fastNLP is a modular and extensible natural language processing (NLP) framework designed to simplify projects by reducing the amount of boilerplate code required for tasks such as data preprocessing and training loops. Key features: - Data preprocessing using apply functions - Customizable training loops - Compatibility with various deep learning frameworks Tech stack: Python Why it matters: fastNLP aims to streamline projects by reducing the amount of repetitive code needed for tasks like data preprocessing and training loops. It provides a flexible and extensible framework that can be easily integrated with different deep learning frameworks, making it a valuable tool for developers working on NLP projects.

Python 3,136 stars

ml-tooling/opyrator

🪄 Turns your machine learning code into microservices with web API, interactive GUI, and more.

Overview: Opyrator is a tool that converts Python functions into production-ready microservices, providing web APIs, interactive GUIs, and more. Key features: - Auto-generated HTTP API based on FastAPI - Auto-generated Web UI based on Streamlit - Save and share as self-contained executable file or Docker image - Reuse pre-defined components and combine with existing Opyrators - Deploy and scale for production usage Tech stack: Python, FastAPI, Streamlit, Pydantic Why it matters: Opyrator simplifies the process of turning Python code into production-ready microservices, making it easier to share and deploy machine learning models or any Python function.

Unknown 3,136 stars

zjunlp/LLMAgentPapers

Must-read Papers on LLM Agents.

### Overview: This repository compiles a collection of essential papers on large language model agents, providing a comprehensive guide for developers and researchers interested in the field of AI and natural language processing. ### Key features: - **Agents Overview**: A detailed introduction to the concept of agents in the context of large language models. - **Agent Types**: Explains various types of agents, including personality, memory, planning, tool use, and RL training. - **Multiple Agents**: Discusses communication between multiple agents in task-oriented and casual conversations. - **Applications**: Explores real-world applications of large language model agents. - **Frameworks**: Provides a framework for understanding and implementing these agents. - **Tools**: Lists various types of tools used in the development of these agents. - **Contributors**: Acknowledges the contributions of multiple researchers and developers. ### Tech stack: The repository is built using Markdown and GitHub Pages. ### Why it matters: This collection serves as a valuable resource for anyone looking to delve into the latest research and applications of large language model agents, offering insights into their capabilities and potential applications.

JavaScript 3,135 stars

SynkraAI/aiox-core

Synkra AIOS: AI-Orchestrated System for Full Stack Development - Core Framework v4.0

### Overview: A Synkra AIOS: AI-Orchestrated System for Full Stack Development - Core Framework v4.0 is a comprehensive framework designed to streamline full-stack development using artificial intelligence. This framework is built on a robust tech stack, ensuring a seamless and efficient development process. ### Key features: 1. **Full Stack Development**: Automates the entire development process from ideation to deployment. 2. **AI Orchestration**: Integrates AI agents to automate tasks and enhance development workflows. 3. **CLI First**: Employs a command-line interface for easy integration and management. ### Tech stack: JavaScript ### Why it matters: This framework provides a powerful toolset for developers, enabling them to focus on their core competencies while the AI orchestrates the development process, ensuring faster, more efficient, and more reliable software solutions.

CSS 3,135 stars

handong1587/handong1587.github.io

Overview: This project is a customizable GitHub blog theme based on the fork of [zJiaJun](https://github.com/zJiaJun)'s theme. It is designed to help developers organize their projects, papers, and reading/study notes in a visually appealing way. Key features: - Customizable design - Organizes content effectively - Easy to use and modify Tech stack: CSS Why it matters: This theme is useful for developers looking to enhance their GitHub profile and keep their projects, papers, and notes organized in a visually appealing manner.

Python 3,133 stars

open-mmlab/mmskeleton

A OpenMMLAB toolbox for human pose estimation, skeleton-based action recognition, and action synthesis.

Overview: MMSkeleton is an open-source toolbox designed for skeleton-based human understanding, part of the open-mmlab project focused on Multimedia Laboratory, CUHK. It supports multiple tasks such as 2D pose estimation, skeleton-based action recognition, and custom dataset creation. Key features: - Supports multiple tasks including 2D pose estimation and skeleton-based action recognition. - Provides a flexible framework for extensibility and scalability. - Supports custom dataset creation and application development. Tech stack: Python. Why it matters: MMSkeleton is essential for researchers and developers working on human pose estimation, action recognition, and skeleton-based applications, offering a comprehensive solution for various tasks in the field of human understanding. Repository: open-mmlab/mmskeleton Primary language: Python GitHub stars: 3133 GitHub description: A OpenMMLab toolbox for human pose estimation, skeleton-based action recognition, and action synthesis. README excerpt: # MMSkeleton ## Introduction MMSkeleton is an open source toolbox for skeleton-based human understanding. It is a part of the [open-mmlab](https://github.com/open-mmlab) project in the charge of [Multimedia Laboratory, CUHK](http://mmlab.ie.cuhk.edu.hk/). MMSkeleton is developed on our research project [ST-GCN](https://github.com/yysijie/st-gcn/blob/master/OLD_README.md). <p align="center"> <img src="demo/recognition/demo_video.gif", width="700"> </p>

Python 3,131 stars

bfortuner/ml-glossary

Machine learning glossary

### Overview: This repository contains a comprehensive glossary of machine learning concepts, designed to be accessible and easy to understand. The glossary is structured to include concise explanations, citations, visuals, code snippets, and equations, making it an excellent resource for developers looking to enhance their machine learning knowledge. ### Key Features: - **Machine Learning Concepts**: Detailed explanations of various machine learning topics. - **Citations**: References to relevant papers and tutorials. - **Visuals**: Diagrams, charts, and animations to illustrate concepts. - **Code Snippets**: Python and NumPy code examples. - **Equations**: Formatted with LaTeX for mathematical expressions. - **Interactive Examples**: Quickstart guides for Jupyter notebooks. ### Tech Stack: Python ### Why It Matters: This glossary serves as a valuable resource for developers, providing a concise and accessible introduction to machine learning concepts. By including visuals, code examples, and equations, it makes complex ideas more understandable and practical. The glossary is designed to be a starting point for rapid prototyping and can be enhanced over time with contributions from other readers. ### GitHub Stars: 3131 ### How To Contribute: 1. **Clone Repo**: `git clone https://github.com/bfortuner/ml-glossary.git` 2. **Install Dependencies**: `pip install sphinx sphinx-autobuild sphinx_rtd_theme recommonmark` 3. **Preview Changes**: `cd ml-glossary; cd docs; make html` 4. **Submit Pull Request**: `https://help.github.com/articles

Python 3,131 stars

fbdesignpro/sweetviz

Visualize and compare datasets, target values and associations, with one line of code.

### Overview: Sweetviz is a Python library designed to streamline Exploratory Data Analysis (EDA) by generating visually appealing reports in just two lines of code. It focuses on quickly visualizing target values and comparing datasets, making it an invaluable tool for data scientists and analysts. ### Key Features: - **Target Analysis**: Shows how a target value (e.g., "Survived" in the Titanic dataset) relates to other features. - **Visualize and Compare**: Distinct datasets (e.g., training vs test data) and intra-set characteristics (e.g., male vs female). - **Mixed-Type Associations**: Seamlessly integrates associations for numerical (Pearson's correlation), categorical (uncertainty coefficient), and categorical-numerical (correlation ratio) datatypes. - **Type Inference**: Automatically detects numerical, categorical, and text features, with optional manual overrides. - **Summary Information**: Type, unique values, missing values, duplicate rows, most frequent values. - **Numerical Analysis**: Min/max/range, quartiles, mean, mode, standard deviation, sum, median absolute deviation, coefficient of variation, kurtosis, skewness. ### Tech Stack: Python ### Why it Matters: Sweetviz simplifies EDA by providing a quick and effective way to analyze datasets and target values, making complex data analysis tasks more accessible and efficient. Its ability to handle mixed-type associations and automatic feature detection makes it a valuable tool for data scientists and analysts alike. ### Repository: fbdesignpro/sweetviz ### GitHub Stars: 3131 ### GitHub

Unknown 3,131 stars

zzw922cn/awesome-speech-recognition-speech-synthesis-papers

Automatic Speech Recognition (ASR), Speaker Verification, Speech Synthesis, Text-to-Speech (TTS), Language Modelling, Singing Voice Synthesis (SVS), Voice Conversion (VC)

### Overview: This repository contains a collection of papers on various aspects of speech recognition, synthesis, and related technologies. ### Key features: - **Automatic Speech Recognition (ASR):** Papers on techniques for converting spoken words into text. - **Speaker Verification:** Papers on methods for verifying the identity of speakers. - **Speech Synthesis (TTS):** Papers on generating spoken text. - **Language Modelling:** Papers on models that predict the probability of words in a sentence. - **Confidence Estimates:** Papers on methods for estimating the confidence of speech recognition results. - **Music Modelling:** Papers on generating music from text or speech. - **Voice Conversion (VC):** Papers on converting one voice into another. - **Interesting papers:** Papers that are particularly interesting or innovative. ### Tech stack: Python, PyTorch, TensorFlow, and other machine learning frameworks. ### Why it matters: This repository provides a comprehensive overview of the latest research in speech technology, including ASR, speaker verification, speech synthesis, language modeling, confidence estimation, music modeling, voice conversion, and more. It serves as a valuable resource for developers and researchers interested in advancing the field of speech technology.

Python 3,130 stars

chenyme/Chenyme-AAVT

这是一个全自动（音频）视频翻译项目。利用Whisper识别声音，AI大模型翻译字幕，最后合并字幕视频，生成翻译后的视频。

### Overview: Chenyme-AAVT 是一个全自动（音频）视频翻译项目，利用 Whisper 识别声音，AI 大模型翻译字幕，最后合并字幕视频生成翻译后的视频。项目旨在提供简便高效且免费的媒体识别与翻译自动化流程。 ### Key features: - **音频识别**: 识别音频中的声音并翻译字幕。 - **视频识别**: 识别视频中的声音并翻译字幕。 - **图文博客**: 生成图文内容。 - **字幕翻译**: 翻译字幕。 - **声音模拟**: 模拟声音。 - **实时语音翻译**: 实时翻译语音。 - **个性化字幕**: 生成个性化字幕。 - **多格式字幕**: 支持多种字幕格式。 - **视频中文配音**: 为视频添加中文配音。 - **口型校对**: 校正口型。 - **视频预览**: 实时预览视频。 - **自动化翻译**: 自动化翻译和二次修改字幕。 ### Tech stack: Python ### Why it matters: Chenyme-AAVT 是一个全自动（音频）视频翻译项目，旨在提供简便高效且免费的媒体识别与翻译自动化流程。它支持多种语言识别和翻译，以及生成图文和实时语音翻译等功能，为用户提供便捷的媒体处理体验。

JavaScript 3,130 stars

edenfunf/reelmimic

Show it a video you love. Get a new video in the same style. An AI crew (Claude Code or Codex) plans, builds and reviews it with you.

### Overview: ReelMimic is an AI-powered tool designed to help users create their own videos by taking inspiration from existing videos. It breaks down the reference video, shows you the plan, and then splits the work among multiple AI agents to ensure a high-quality final product. ### Key Features: - **Reference Breakdown:** The tool analyzes the reference video to understand the rhythm, shot lengths, transitions, colors, framing, and camera moves. - **Storyboard and Style Frames:** It provides a storyboard and style frames for you to review and approve before proceeding. - **Multiple AI Agents:** Up to 6 AI agents work on different parts of the video simultaneously, ensuring a smooth and professional production process. - **Fixes and Reviews:** Every change made is accompanied by before and after screenshots, and reviewers check them to ensure accuracy. - **Real-Time Feedback:** You can see what each AI agent is thinking and what it's looking at, as well as the full log of the production process. - **Comment and Notes:** You can leave comments on the video at any point, and it can be sent to all reviewers at once. - **Versatile Styles:** You can choose from three languages: English, Chinese (繁體中文), and Chinese (简体中文). - **2D Only:** The tool is designed for 2D videos. ### Tech Stack: JavaScript ### Why It Matters: ReelMimic is a powerful tool for anyone looking to create their own videos, especially those who want to maintain the style and feel of an existing video while making it

Unknown 3,128 stars

jingyi0000/VLM_survey

Collection of AWESOME vision-language models for vision tasks

### Overview: This repository is a collection of **Vision Language Models for Vision Tasks** (VLMs) surveys, focusing on various visual recognition tasks such as image classification, object detection, and semantic segmentation. The survey is based on the paper "Vision-Language Models for Vision Tasks: A Survey" published in IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2024. ### Key features: - Comprehensive coverage of VLM studies across different visual recognition tasks. - Detailed analysis and comparison of various VLM models and techniques. - Links to relevant papers, code, and conferences for further exploration. ### Tech stack: - Python - Markdown - LaTeX ### Why it matters: This survey serves as a valuable resource for researchers and practitioners interested in the latest advancements in VLMs, providing a clear overview of the state of the art in vision-language models for various vision tasks.

C 3,127 stars

HeyWillow/willow

Open source, local, and self-hosted Amazon Echo/Google Home competitive Voice Assistant alternative

Overview: Willow is an open-source, self-hosted alternative to Amazon Echo/Google Home, offering a competitive voice assistant experience. Key features: - Self-hosted inference server - Lightning-fast language inference tasks - Supports Willow and other applications, including STT, TTS, LLM, and more - Enables WebRTC integration Tech stack: C Why it matters: Willow provides a reliable, local voice assistant solution for developers and enthusiasts, offering a competitive alternative to popular voice assistants while maintaining the ability to integrate with existing applications and technologies.

TypeScript 3,127 stars

mintlify/writer

✍️ AI powered documentation writer

Overview: Mintlify Writer is an AI-powered documentation writer that simplifies the process of generating documentation for various programming languages. Key features: - Supports TypeScript - Highlights code for documentation - Compatible with Python, JavaScript, TypeScript, JSX/TSX, C/C++, PHP, Java, C#, Ruby, Rust, and more - Supports Markdown, JSDoc, reST, NumPy, DocBlock, Doxygen, Javadoc, GoDoc, XML, and Google formats Tech stack: TypeScript Why it matters: Mintlify Writer automates the documentation generation process, making it easier for developers to maintain and update documentation for their projects. Repository: mintlify/writer GitHub stars: 3127

Python 3,126 stars

docarray/docarray

Represent, send, store and search multimodal data

Overview: DocArray is a Python library designed for the representation, transmission, storage, and retrieval of multimodal data, making it ideal for the development of multimodal AI applications. Key features: - Native support for NumPy, PyTorch, TensorFlow, and JAX - Pydantic-based framework compatibility with FastAPI and Jina - Support for vector databases like Weaviate, Qdrant, ElasticSearch, Redis, MongoDB Atlas, and HNSWLib - Data transmission as JSON over HTTP or Protobuf over gRPC Tech stack: Python Why it matters: DocArray simplifies the integration of multimodal data into AI applications, making it easier to develop and deploy models that can handle various types of data, including images, text, audio, and more.

Python 3,125 stars

robert-mcdermott/ai-knowledge-graph

AI Powered Knowledge Graph Generator

### Overview: This project is an AI-powered Knowledge Graph Generator that converts unstructured text documents into interactive knowledge graphs using an LLM. It supports various input formats and provides features like extraction, entity standardization, conservative inference, and export options. ### Key Features: - **Any Text Input**: Supports `.txt`, `.md`, `.rst`, `.pdf`, and `.docx` files or directories. - **Typed Knowledge Extraction**: Extracts Subject-Predicate-Object triples with entity types and keeps the sentence context. - **Entity Standardization**: Merges case, stop-word, and plural variants with an optional LLM pass. - **Conservative, Traceable Inference**: Bridges isolated parts of the graph and adds well-known relationships between central entities. - **Exports**: JSON, CSV, GraphML for Gephi/yEd/Cytoscape, and a Cypher script for Neo4j. - **Chat with the Graph** (optional `graph-chat` command): Grounded, cited answers from the generated graph. - **Local Web Interface** (optional `graph-serve` command): Ingest, browse, explore, and ask questions in the browser. - **Interactive Explorer**: A single self-contained HTML file with search, click-to-highlight, relationships panel, named communities, entity-type filters, shortest-path finder, exports, and light/dark themes. - **Robust LLM Client**: Truncation detection, retries with back-off, automatic `max_completion_tokens` fallback, environment-variable API keys, and on-disk response cache. - **Works with

Jupyter Notebook 3,122 stars

decodingai-magazine/second-brain-ai-assistant-course

Learn to build your Second Brain AI assistant with LLMs, agents, RAG, fine-tuning, LLMOps and AI systems techniques.

### Overview: Learn to build your Second Brain AI assistant with LLMs, agents, RAG, fine-tuning, LLMOps and AI systems techniques. ### Key features: - Build an agentic RAG system powered by your Second Brain - Design production-ready LLM architectures - Apply LLMOps and software engineering best practices - Fine-tune and deploy LLMs - Use industry tools: OpenAI, Hugging Face, MongoDB, ZenML, Opik, Comet, Unsloth, and more ### Tech stack: Jupyter Notebook ### Why it matters: Access the collective wisdom of your own mind, build an advanced RAG and LLM system using LLMOps and ML systems best practices.

Unknown 3,118 stars

luspr/awesome-ml-courses

Awesome free machine learning and AI courses with video lectures.

### Overview: This repository contains a curated list of free machine learning and artificial intelligence courses with video lectures, designed to help beginners and intermediate learners get started in the field. The courses cover a wide range of topics, from introductory lectures to advanced techniques, and include detailed lecture notes, additional readings, and assignments. ### Key Features: - [Machine Learning (Stanford CS229)](https://www.youtube.com/playlist?list=PLoROMvodv4rMiGQp3WXShtMGgzqpfVfbU) - [Convolutional Neural Networks for Visual Recognition (Stanford CS231n)](https://www.youtube.com/playlist?list=PL3FW7Lu3i5JvHM8ljYj-zLfQRF3EO8sYv) - [Introduction to Artificial Intelligence (UC Berkeley CS188)](https://www.youtube.com/playlist?list=PL7k0r4t5c108AZRwfW-FhnkZ0sCKBChLH) - [Applied Machine Learning 2020 (Columbia)](https://www.youtube.com/playlist?list=PL_pVmAaAnxIRnSw6wiCpSvshFyCREZmlM) - [Introduction to Reinforcement learning with David Silver (DeepMind)](https://www.youtube.com/playlist?list=PLqYmG7hTraZBiG_XpjnPrSNw-1XQaM_g