Top GitHub Projects of 04/13/2023: Discovering the Fastest Rising Stars of Today!
Projects that have already appeared in previous Github ranking lists will not display details, please refer to previous lists for information.
Rank #1 reworkd/AgentGPT
https://github.com/reworkd/AgentGPT
🤖 Assemble, configure, and deploy autonomous AI Agents in your browser.
Language: TypeScript
Stars: 7,178(3,433 stars today) Forks:1,372
— — — — — — — — — — — — — — — —
Rank #2 biobootloader/wolverine
https://github.com/biobootloader/wolverine
Language: Python
Stars: 2,730(1,091 stars today) Forks:274
— — — — — — — — — — — — — — — —
Rank #3 ravenscroftj/turbopilot
https://github.com/ravenscroftj/turbopilot
Turbopilot is an open source large-language-model based code completion engine that runs locally on CPU
Language: Python
Stars: 2,228(745 stars today) Forks:65
— — — — — — — — — — — — — — — —
Rank #4 eumemic/ai-legion
https://github.com/eumemic/ai-legion
An LLM-powered autonomous agent platform
Language: TypeScript
Stars: 486(125 stars today) Forks:48
The AI Legion project is an autonomous agent platform powered by LLM (Large Language Model) technology. It provides a framework for autonomous agents to work together to accomplish tasks. The project requires Node 10 or higher and involves setting up secrets in the `.env` file, such as the OpenAI API key and Google Custom Search API key, to allow the agent to search the web. The program can be started with `npm run start [# of agents] [gpt-3.5-turbo|gpt-4]`, and the user can interact with the agents through the console. The agents may make mistakes initially as they learn to operate themselves, but they generally learn from their mistakes and recover. Each agent stores its state under the `.store` directory, which allows for selective wiping of state between runs and effective replay of moments for debugging. This project has potential applications in fields such as automation, natural language processing, and machine learning. It can be used commercially to develop autonomous agents that can work together to accomplish complex tasks, such as customer service, data analysis, and decision-making.
— — — — — — — — — — — — — — — —
Rank #5 Torantulino/Auto-GPT
https://github.com/Torantulino/Auto-GPT
An experimental open-source attempt to make GPT-4 fully autonomous.
Language: Python
Stars: 46,032(12,928 stars today) Forks:6,069
— — — — — — — — — — — — — — — —
Rank #6 ohmplatform/FreedomGPT
https://github.com/ohmplatform/FreedomGPT
This codebase is for a React and Electron-based app that executes the FreedomGPT LLM locally (offline and private) on Mac and Windows using a chat-based interface (based on Alpaca Lora)
Language: TypeScript
Stars: 730(349 stars today) Forks:75
— — — — — — — — — — — — — — — —
Rank #7 microsoft/DeepSpeed
https://github.com/microsoft/DeepSpeed
DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.
Language: Python
Stars: 15,021(3,616 stars today) Forks:1,743
DeepSpeed is a deep learning optimization software suite that provides extreme speed and scale for both training and inference of language models such as MT-530B and BLOOM. It offers a range of features, including automatic tensor parallelism, data efficiency, and ZeRO-Inference, which democratizes massive model inference. DeepSpeed enables the training and inference of dense or sparse models with billions or trillions of parameters, making it applicable in fields such as natural language processing, computer vision, and speech recognition. Commercial applications of DeepSpeed include developing high-performance language models for customer service, personal assistants, and chatbots, as well as improving the efficiency and accuracy of data analysis and decision-making in various industries. The latest news on the DeepSpeed project includes the release of DeepSpeed Chat, which offers easy, fast, and affordable RLHF training of ChatGPT-like models at all scales, and the development of VL-MoE, which scales large-scale generative mixture-of-expert multimodal models.
— — — — — — — — — — — — — — — —
Rank #8 fudan-zvg/Semantic-Segment-Anything
https://github.com/fudan-zvg/Semantic-Segment-Anything
Automated dense category annotation engine that serves as the initial semantic labeling for the Segment Anything dataset (SA-1B).
Language: Python
Stars: 433(66 stars today) Forks:16
The Semantic Segment Anything (SSA) project is an automated annotation engine that enhances the Segment Anything dataset (SA-1B) with dense category annotation. The project uses a combination of close-set segmentation and open-vocabulary segmentation to produce satisfactory labeling for most samples and has the capability to provide more detailed annotations using image caption method. SSA fills the gap in SA-1B’s limited fine-grained semantic labeling and significantly reduces the need for manual annotation and associated costs. It has the potential to serve as a foundation for training large-scale visual perception models and more fine-grained CLIP models. SSA can be applied in fields such as computer vision, natural language processing, and machine learning. Commercial applications of SSA include developing image segmentation models for object detection, autonomous driving, and medical imaging, as well as improving the accuracy and efficiency of data analysis in various industries. The SSA engine consists of three components: a close-set semantic segmentor, an open-vocabulary classifier, and a dense category annotation engine. The combination of fine image segmentation annotations of SA-1B with the rich semantic annotations provided by advanced models makes SSA a valuable tool for generating densely categorized image segmentation datasets.
— — — — — — — — — — — — — — — —
Rank #9 continue-revolution/sd-webui-segment-anything
https://github.com/continue-revolution/sd-webui-segment-anything
Segment Anything for Stable Diffusion Webui
Language: Python
Stars: 398(93 stars today) Forks:17
The Segment Anything for Stable Diffusion WebUI project is an extension that enables users of the stable diffusion webui to use segment anything for stable diffusion inpainting. The project includes features such as mask expansion and preview version of GroundingDINO support. The project plans to support text->detection->segmentation from Grounded Segment Anything, API support, automatic mask generation for hierarchical image segmentation and SD animation, semantic segmentation for batch process, and connection to ControlNet inpainting and segmentation. This project can be applied in fields such as computer vision, image processing, and machine learning. Commercial applications of this project include developing image inpainting models for photo editing and restoration, as well as improving the efficiency and accuracy of data analysis and decision-making in various industries. The project is designed to work with the stable diffusion webui and segment anything model checkpoints. The extension can be downloaded to the sd-webui/extensions directory, and the segment-anything model can be downloaded to the sd-webui/models/sam directory. This project aims to provide stable diffusion webui users with a more comprehensive and efficient image inpainting tool.
— — — — — — — — — — — — — — — —
Rank #10 databrickslabs/dolly
https://github.com/databrickslabs/dolly
Databricks’ Dolly, a large language model trained on the Databricks Machine Learning Platform
Language: Python
Stars: 5,861(1,233 stars today) Forks:561
— — — — — — — — — — — — — — — —
Rank #11 microsoft/DeepSpeedExamples
https://github.com/microsoft/DeepSpeedExamples
Example models using DeepSpeed
Language: Python
Stars: 1,671(725 stars today) Forks:306
— — — — — — — — — — — — — — — —
Rank #12 kevmo314/magic-copy
https://github.com/kevmo314/magic-copy
Magic Copy is a Chrome extension that uses Meta’s Segment Anything Model to extract a foreground object from an image and copy it to the clipboard.
Language: TypeScript
Stars: 869(334 stars today) Forks:47
— — — — — — — — — — — — — — — —
Rank #13 practical-tutorials/project-based-learning
https://github.com/practical-tutorials/project-based-learning
Curated list of project-based tutorials
Language:
Stars: Star(328 stars today) Forks:13,723
The Project Based Learning project is a collection of programming tutorials aimed at aspiring software developers who want to learn how to build an application from scratch. The tutorials cover a wide range of programming languages, including C#, C/C++, Clojure, Dart, Elixir, Erlang, F#, Go, Haskell, HTML/CSS, Java, JavaScript, Kotlin, Lua, OCaml, PHP, Python, R, Ruby, Rust, Scala, and Swift. Each tutorial may involve multiple technologies and languages. The tutorials cover various topics, including building an interpreter, writing a shell, building a file system, creating an operating system from scratch, implementing a key-value store, and writing a hash table. These tutorials can be applied in fields such as software development, computer science, and information technology. Commercial applications of this project include developing software applications, web applications, mobile applications, and games. The tutorials provide a valuable resource for aspiring software developers to learn new programming languages and technologies and to build their skills in software development. The project is open source and contributions are welcome from the community.
— — — — — — — — — — — — — — — —
Rank #14 hemansnation/God-Level-Data-Science-ML-Full-Stack
https://github.com/hemansnation/God-Level-Data-Science-ML-Full-Stack
This roadmap contains 16 Chapters that can be completed in 8 months, whether you are a fresher in the field or an experienced professional who wants to transition into Data Science.
Language: Jupyter Notebook
Stars: 1,770(344 stars today) Forks:356
— — — — — — — — — — — — — — — —
Rank #15 agiresearch/OpenAGI
https://github.com/agiresearch/OpenAGI
OpenAGI: When LLM Meets Domain Experts
Language: Jupyter Notebook
Stars: 604(240 stars today) Forks:35
The OpenAGI project is an open-source Artificial General Intelligence (AGI) research platform that aims to equip large language models (LLMs) with the capability to harness various domain-specific expert models for complex task-solving. The project presents a Reinforcement Learning from Task Feedback (RLTF) mechanism, which uses the task-solving result as feedback to improve the LLM’s task-solving ability. The LLM is responsible for synthesizing various external models for solving complex tasks, while RLTF provides feedback to improve its task-solving ability, enabling a feedback loop for self-improving AI. The project formulates complex tasks as natural language queries, serving as input to the LLM. The project includes task-specific datasets, evaluation metrics, and a diverse range of extensible models. The OpenAGI project can be applied in fields such as artificial intelligence, machine learning, and natural language processing. Commercial applications of this project include developing self-improving AI agents for various industries, such as finance, healthcare, and transportation. The project is open source, and the code, benchmark, and evaluation methods are available for the community to use and improve upon. The project requires Python 3.9.16 and PyTorch 1.12.1, and the preprocessed data can be downloaded from a Google Drive link provided in the project. The OpenAGI project represents a promising approach towards AGI, and the community’s contributions and suggestions are welcome.
— — — — — — — — — — — — — — — —
Rank #16 anuragxel/salt
https://github.com/anuragxel/salt
Segment Anything Labelling Tool
Language: Python
Stars: 454(241 stars today) Forks:30
— — — — — — — — — — — — — — — —
Rank #17 discordjs/discord.js
https://github.com/discordjs/discord.js
A powerful JavaScript library for interacting with the Discord API
Language: TypeScript
Stars: 22,630(209 stars today) Forks:3,801
— — — — — — — — — — — — — — — —
Rank #18 Tencent/libpag
https://github.com/Tencent/libpag
The official rendering library for PAG (Portable Animated Graphics) files that renders After Effects animations natively across multiple platforms.
Language: C++
Stars: 2,909(49 stars today) Forks:313
The libpag project is a real-time rendering library for Portable Animated Graphics (PAG) files that renders both vector-based and raster-based animations across most platforms, such as iOS, Android, macOS, Windows, Linux, and Web. PAG is an open-source file format for recording animations, which can be created and exported from Adobe After Effects with the PAGExporter plugin and previewed in the PAGViewer app. The libpag project is being used by 40+ Tencent apps, such as WeChat, Mobile QQ, Honor of Kings Mobile Game, Tencent Video, QQ Music, and so on, reaching hundreds of millions of users. The project offers several advantages, including highly efficient file format, support for all After Effects (AE) features, measurable performance, and runtime editable animations. The project can be applied in fields such as animation, multimedia, and entertainment. Commercial applications of this project include developing multimedia applications, video templates, and interactive user interfaces for various industries, such as advertising, gaming, and education. The project is open source, and the code, benchmark, and evaluation methods are available for the community to use and improve upon. The project requires iOS 9.0 or later, Android 4.4 or later, macOS 10.13 or later, Windows 7.0 or later, or Chrome.
— — — — — — — — — — — — — — — —
Rank #19 shadcn/taxonomy
https://github.com/shadcn/taxonomy
An open source application built using the new router, server components and everything new in Next.js 13.
Language: TypeScript
Stars: 7,664(770 stars today) Forks:692
— — — — — — — — — — — — — — — —
Rank #20 ItsPi3141/alpaca-electron
https://github.com/ItsPi3141/alpaca-electron
An even simpler way to run Alpaca
Language: JavaScript
Stars: 443(78 stars today) Forks:62
Alpaca Electron is a software tool that enables users to chat with Alpaca AI models without the need for command line or compiling. It is built to be user-friendly and can run on any computer without requiring an expensive graphics card. Alpaca Electron uses a compact and efficient backend called alpaca.cpp, which makes it easy to download and install. The tool supports Windows, MacOS, and Linux, and it can be used for a wide range of applications, including chatbots, virtual assistants, and customer service. It also includes features like context memory and chat history, with plans for integration with Stable Diffusion in the future. Alpaca Electron is a powerful tool with many commercial applications, including e-commerce, healthcare, and finance, where it can be used to automate customer service and provide personalized support.
— — — — — — — — — — — — — — — —
Rank #21 IDEA-Research/Grounded-Segment-Anything
https://github.com/IDEA-Research/Grounded-Segment-Anything
Marrying Grounding DINO with Segment Anything & Stable Diffusion & BLIP & Whisper — Automatically Detect , Segment and Generate Anything with Image, Text, and Speech Inputs
Language: Jupyter Notebook
Stars: 6,017(573 stars today) Forks:434
— — — — — — — — — — — — — — — —
Rank #22 vinta/awesome-python
https://github.com/vinta/awesome-python
A curated list of awesome Python frameworks, libraries, software and resources
Language: Python
Stars: 163,899(293 stars today) Forks:22,933
— — — — — — — — — — — — — — — —
Rank #23 bfeber/HLA-NoVR
https://github.com/bfeber/HLA-NoVR
NO VR Script for Half-Life: Alyx
Language: Lua
Stars: 352(81 stars today) Forks:21
— — — — — — — — — — — — — — — —
Rank #24 xx025/carrot
https://github.com/xx025/carrot
Free ChatGPT Site List 这儿为你准备了众多免费好用的ChatGPT镜像站点，当前100+站点
Language:
Stars: Star(185 stars today) Forks:628
— — — — — — — — — — — — — — — —
Rank #25 vivo/MoonBox
https://github.com/vivo/MoonBox
月光宝盒：无侵入的流量录制与回放平台 A server-side traffic capture and replay platform with noninvasive
Language: Java
Stars: 276(16 stars today) Forks:55
Moonbox is a traffic replay platform developed based on JVM-Sandbox ecosystem and jvm-sandbox-repeater. It is a non-invasive online traffic recording and playback platform that provides a wide range of features, including automated testing, online problem tracking, and business monitoring. Moonbox uses the SPI design of jvm-sandbox-repeater and provides numerous commonly used plugins, as well as data statistics and storage capabilities. It can be applied in various fields such as automated testing, online problem tracking, and business monitoring. Moonbox enables users to record and store request input and output parameters, downstream RPC, DB, and cache sequences, and replay the recorded data to re-execute one or multiple requests, or to perform MOCK on specific downstream nodes. Moonbox also provides features such as noise reduction, comparison configuration, and more. The platform can be deployed easily and is compatible with various applications, including http/dubbo. Moonbox has been used by over 100 projects at vivo for two years, and the team plans to open-source more features in the future, such as more plugins, MySQL data storage, C++ traffic recording and playback, Docker platform deployment, and more.
— — — — — — — — — — — — — — — —
