Top GitHub Projects of 04/19/2023: Discovering the Fastest Rising Stars of Today

Adair Lee
10 min readApr 19, 2023

--

Top 25 Fastest Growing GitHub Projects

Projects that have already appeared in previous Github ranking lists will not display details, please refer to previous lists for information.

Rank #1 Vision-CAIR/MiniGPT-4
https://github.com/Vision-CAIR/MiniGPT-4
MiniGPT-4: Enhancing Vision-language Understanding with Advanced Large Language Models
Language: Python
Stars: 6,783(2,390 stars today) Forks:596
MiniGPT-4 is an advanced language model that enhances vision-language understanding. The model aligns a frozen visual encoder with a frozen large language model, Vicuna, using just one projection layer. MiniGPT-4 is trained in two stages, the first traditional pre-training stage is trained using roughly 5 million aligned image-text pairs in 10 hours, and the second fine-tuning stage is trained on a small yet high-quality dataset to significantly improve its generation reliability and overall usability. The model yields many emerging vision-language capabilities similar to those demonstrated in GPT-4. MiniGPT-4 can be applied in various fields, including computer vision, natural language processing, and artificial intelligence. The commercial applications of MiniGPT-4 include image captioning, visual question answering, and language-based image retrieval systems.
— — — — — — — — — — — — — — — —

Rank #2 facebookresearch/dinov2
https://github.com/facebookresearch/dinov2
PyTorch code and models for the DINOv2 self-supervised learning method.
Language: Python
Stars: 2,480(801 stars today) Forks:116
DINOv2 is a self-supervised learning framework for learning robust visual features without supervision. The project is developed by Meta AI Research and Facebook AI Research. The models produced by DINOv2 can be directly employed with classifiers as simple as linear layers on a variety of computer vision tasks. The models were pretrained on a dataset of 142 million images without using any labels or annotations. DINOv2 can be applied in various fields, including object recognition, image classification, and image retrieval. The commercial applications of DINOv2 include developing image-based search engines, content-based image retrieval systems, and image recognition systems for autonomous vehicles. The pretrained models for DINOv2 are available for download via PyTorch Hub.
— — — — — — — — — — — — — — — —

Rank #3 muellerberndt/micro-gpt
https://github.com/muellerberndt/micro-gpt
A minimal generic autonomous agent based on GPT3.5/4. Can analyze stock prices, perform network security tests, create art, and order pizza.
Language: Python
Stars: 908(457 stars today) Forks:80
— — — — — — — — — — — — — — — —

Rank #4 houseofsecrets/SdPaint
https://github.com/houseofsecrets/SdPaint
Stable Diffusion Painting
Language: Python
Stars: 750(186 stars today) Forks:50
SdPaint is a simple Python script that lets you paint on a canvas and sends that image every stroke to the automatic1111 API and updates the canvas when the image is generated. The project is designed to work with the automatic1111 webui in API mode, which can be used for various applications such as image generation, text-to-image synthesis, and image manipulation. SdPaint can be applied in various fields, including art, design, and computer graphics. The commercial applications of SdPaint include developing image generation tools, creating custom images for websites and applications, and designing logos and graphics for businesses. The project is open-source and available under the MIT license. Contributions and pull requests are welcome.
— — — — — — — — — — — — — — — —

Rank #5 geekyutao/Inpaint-Anything
https://github.com/geekyutao/Inpaint-Anything
Inpaint anything using Segment Anything and inpainting models.
Language: Jupyter Notebook
Stars: 1,019(266 stars today) Forks:57
Inpaint Anything is a computer vision project that allows users to remove any object from an image by clicking on it and then fill the object with any desired content or replace the background of it arbitrarily. The project is developed by researchers from the University of Science and Technology of China and Eastern Institute for Advanced Study. Inpaint Anything is powered by powerful vision models, such as SAM, LaMa, and Stable Diffusion (SD). The project can be applied in various fields, including image editing, content creation, and graphic design. The commercial applications of Inpaint Anything include developing image editing software, creating custom images for websites and applications, and designing logos and graphics for businesses. Inpaint Anything is open-source and can be installed using Python pip. The project supports any aspect ratio and 2K resolution. The demo website for Inpaint Anything is coming soon.
— — — — — — — — — — — — — — — —

Rank #6 togethercomputer/RedPajama-Data
https://github.com/togethercomputer/RedPajama-Data
The RedPajama-Data repository contains code for preparing large datasets for training large language models.
Language: Python
Stars: 999(396 stars today) Forks:60
RedPajama-Data is an open-source project that provides a recipe to reproduce the LLaMA training dataset. The dataset consists of the following token counts: Commoncrawl (878 Billion), C4 (175 Billion), GitHub (59 Billion), Books (26 Billion), ArXiv (28 Billion), Wikipedia (24 Billion), StackExchange (20 Billion), and Total (1.2 Trillion). The project provides all pre-processing scripts and guidelines for data preparation and an example of how to tokenize the dataset using the GPT-NeoX tokenizer. The project also includes a dashboard for exploring a subset of the data using Meerkat. RedPajama-Data can be applied in various fields, including natural language processing, machine learning, and data science. The commercial applications of RedPajama-Data include developing language models, building chatbots, and creating intelligent virtual assistants. The project is open-source and licensed under the Apache 2.0 license. The dataset itself is subject to the licenses of the data subsets used.
— — — — — — — — — — — — — — — —

Rank #7 zilliztech/GPTCache
https://github.com/zilliztech/GPTCache
GPTCache is a library for creating semantic cache to store responses from LLM queries.
Language: Python
Stars: 1,302(191 stars today) Forks:91
GPTCache is a library for creating a semantic cache for large language model (LLM) queries, which can reduce API costs by 10x and boost speed by 100x. The project is developed by Zilliztech and is licensed under the MIT license. GPTCache is designed to tackle the challenge of high expenses related to LLM API calls and slow response times when dealing with a significant number of requests. The project provides a semantic cache for storing LLM responses. GPTCache can be applied in various fields, including natural language processing, machine learning, and data science. The commercial applications of GPTCache include developing language models, building chatbots, and creating intelligent virtual assistants. The project is open-source and can be installed using pip. However, the API may be subject to change at any time, as the project is undergoing swift development.
— — — — — — — — — — — — — — — —

Rank #8 shadcn/ui
https://github.com/shadcn/ui
Beautifully designed components built with Radix UI and Tailwind CSS.
Language: TypeScript
Stars: 12,356(337 stars today) Forks:456
— — — — — — — — — — — — — — — —

Rank #9 yoavbls/pretty-ts-errors
https://github.com/yoavbls/pretty-ts-errors
🔵 Make TypeScript errors prettier and human-readable in VSCode 🎀
Language: TypeScript
Stars: 7,143(2,005 stars today) Forks:34
The “Pretty TypeScript Errors” project is a Visual Studio Code extension that makes TypeScript errors more human-readable and visually appealing. TypeScript errors can become complex and difficult to understand, especially when dealing with complex types. This extension helps developers by providing syntax highlighting with theme colors for types in error messages, a button that leads to the relevant type declaration, and a button that navigates to a detailed explanation of the error on typescript.tv. The project supports Node and Deno TypeScript error reporters, JSDoc type errors, React, Solid, and Qwik errors, and Astro, Svelte, and Vue files when TypeScript is enabled. The project also addresses some challenges, such as the inconsistency of types that are not valid in TypeScript, the inability to syntax highlight types in code blocks, and the difficulty of styling error messages in VSCode markdown. The project can be useful for developers working with TypeScript in various fields, such as web development, software development, and mobile app development. Commercial applications of the project include improving productivity and reducing development time by simplifying and streamlining the error debugging process.
— — — — — — — — — — — — — — — —

Rank #10 liou666/polyglot
https://github.com/liou666/polyglot
🤖️ 桌面端AI语言练习应用
Language: TypeScript
Stars: 445(160 stars today) Forks:43
The “Polyglot” project is a desktop application designed for language practice using AI language models. The project is built using Vue3+TS technology stack and utilizes ChatGPT and Azure AI language models as underlying services. The application aims to provide an easy-to-use language practice platform for users to practice speaking multiple languages. The project features intelligent voice synthesis based on Azure TTS service and intelligent conversation functionality based on ChatGPT service. It also supports dark mode, text translation, and user-defined languages and AI characters. The project can be applied in various fields, such as language learning, language teaching, and language research. Commercial applications of the project include providing language practice and training services for individuals and organizations. The project is open source and welcomes contributions from developers.
— — — — — — — — — — — — — — — —

Rank #11 LAION-AI/Open-Assistant
https://github.com/LAION-AI/Open-Assistant
OpenAssistant is a chat-based assistant that understands tasks, can interact with third-party systems, and retrieve information dynamically to do so.
Language: Python
Stars: 28,116(1,014 stars today) Forks:2,284
— — — — — — — — — — — — — — — —

Rank #12 linyiLYi/street-fighter-ai
https://github.com/linyiLYi/street-fighter-ai
This is an AI agent for Street Fighter II Champion Edition.
Language: Python
Stars: 3,469(512 stars today) Forks:624
— — — — — — — — — — — — — — — —

Rank #13 hax/heshijun_v_360
https://github.com/hax/heshijun_v_360
贺师俊与360的劳动争议诉讼
Language:
Stars: Star(349 stars today) Forks:147
— — — — — — — — — — — — — — — —

Rank #14 mlc-ai/web-llm
https://github.com/mlc-ai/web-llm
Bringing large-language models and chat to web browsers. Everything runs inside the browser with no server support.
Language: Python
Stars: 3,432(420 stars today) Forks:147
— — — — — — — — — — — — — — — —

Rank #15 iamadamdev/bypass-paywalls-chrome
https://github.com/iamadamdev/bypass-paywalls-chrome
Bypass Paywalls web browser extension for Chrome and Firefox.
Language: JavaScript
Stars: 35,887(404 stars today) Forks:2,637
The “Bypass Paywalls” project is a web browser extension that helps users bypass paywalls for selected sites. The project currently supports custom sites on Google Chrome and Microsoft Edge, while Mozilla Firefox does not support custom sites. The project provides users with a list of sites that can bypass paywalls, including Adweek, American Banker, Bloomberg, Boston Globe, Business Insider, Financial Times, and many others. The project can be applied in various fields, such as journalism, research, and education. Commercial applications of the project include providing access to paywalled content for individuals and organizations. However, it should be noted that bypassing paywalls without permission may be illegal and unethical, and users should use the project responsibly and legally.
— — — — — — — — — — — — — — — —

Rank #16 Trendyol/baklava
https://github.com/Trendyol/baklava
Design System with Web Components
Language: TypeScript
Stars: 743(76 stars today) Forks:67
The “Baklava Design System” project is a design system created by Trendyol to provide a consistent UI/UX for app users. The project provides a web implementation of the design system as native web components, making it compatible with various web frameworks such as Vue, React, and Angular. The project aims to provide a UI library that is well-designed and developed to provide the best possible user experience for the users of applications that use Baklava DS. The project can be applied in various fields, such as web development, app development, and UI/UX design. Commercial applications of the project include improving the user experience and consistency of web and app interfaces for businesses and organizations. The project is open source and welcomes contributions from designers and developers to improve the design system.
— — — — — — — — — — — — — — — —

Rank #17 nomic-ai/gpt4all-chat
https://github.com/nomic-ai/gpt4all-chat
gpt4all-j chat
Language: C++
Stars: 582(88 stars today) Forks:55
The “gpt4all-chat” project is a cross-platform GUI application based on the GPT-J language model. The project is built using Qt technology and provides a fast CPU-based inference using ggml for GPT-J based models. The UI is designed to look and feel like a chatty AI, with features such as check for updates, easy installation, and advanced settings for changing temperature, topk, etc. The project can be applied in various fields, such as natural language processing, machine learning, and chatbot development. Commercial applications of the project include providing conversational interfaces for businesses and organizations. The project is open source and welcomes contributions from developers to add new features, such as multi-chat, text to speech, speech to text, syntax highlighting support, and REST API with a built-in web server. The project also provides building and running instructions for both Qt-based GUI and CLI tools.
— — — — — — — — — — — — — — — —

Rank #18 pengzhile/pandora
https://github.com/pengzhile/pandora
潘多拉,一个不只是命令行的ChatGPT。
Language: Python
Stars: 1,333(217 stars today) Forks:212
— — — — — — — — — — — — — — — —

Rank #19 EwingYangs/awesome-open-gpt
https://github.com/EwingYangs/awesome-open-gpt
Collection of Open Source Projects Related to GPT,GPT相关开源项目合集🚀、精选🔥🔥
Language: Python
Stars: 1,120(291 stars today) Forks:105
— — — — — — — — — — — — — — — —

Rank #20 ading2210/poe-api
https://github.com/ading2210/poe-api
A reverse engineered Python API wrapper for Quora’s Poe, which provides free access to ChatGPT, GPT-4, and Claude.
Language: Python
Stars: 648(31 stars today) Forks:80
— — — — — — — — — — — — — — — —

Rank #21 imClumsyPanda/langchain-ChatGLM
https://github.com/imClumsyPanda/langchain-ChatGLM
langchain-ChatGLM, local knowledge based ChatGLM with langchain | 基于本地知识的 ChatGLM 问答
Language: Python
Stars: 2,500(200 stars today) Forks:256
— — — — — — — — — — — — — — — —

Rank #22 zabirauf/AutoGPT.js
https://github.com/zabirauf/AutoGPT.js
Auto-GPT on the browser
Language: TypeScript
Stars: 382(155 stars today) Forks:58
— — — — — — — — — — — — — — — —

Rank #23 Significant-Gravitas/Auto-GPT-Plugins
https://github.com/Significant-Gravitas/Auto-GPT-Plugins
Plugins for Auto-GPT
Language:
Stars: Star(174 stars today) Forks:42
— — — — — — — — — — — — — — — —

Rank #24 OpenBB-finance/OpenBBTerminal
https://github.com/OpenBB-finance/OpenBBTerminal
Investment Research for Everyone, Anywhere.
Language: Python
Stars: 21,013(296 stars today) Forks:2,101
— — — — — — — — — — — — — — — —

Rank #25 facebookresearch/llama
https://github.com/facebookresearch/llama
Inference code for LLaMA models
Language: Python
Stars: 18,866(144 stars today) Forks:3,034
— — — — — — — — — — — — — — — —

--

--

Adair Lee

Experienced full-stack developer proficient in C#, Python, and web development, with 20+ years of Google SEO expertise and successful entrepreneurship.