Hey there! 👋 Welcome to #TrendingTuesday
This week we'll look into the fastest growing machine-learning repos of the week
Our weekly trending lists take into account the historical performance of similar repos and the repo's current growth stage. For this reason, you’ll find some gems here that you won’t see anywhere else 💎
Remember to follow us on Twitter @quine and stay up-to-date with new product announcements 🐦
9️⃣ In the ninth place we have markprompt, a tool for creating GPT-powered prompts from Markdown files in your GitHub repo. It includes a React component for building the prompts, as well as analytics to track user interactions with your docs!
It was brought to you by the team at @motifland and has ~7 issues and ~5 contributors.
8️⃣ Next is ebook-GPT-translator, an OpenAI powered tool that translates text to multiple languages and even converts it to other file formats like PDF, DOCX, EPUB, and MOBI.
It was created by @jesselaunz and has ~15 issues and ~3 contributors.
7️⃣ In number seven we have opencommit, a GPT-powered program for the CLI that allows you to generate impressive commit messages in one second! You can configure it to preface commits with an emoji and post-face them with description of changes.
Really impressive (and useful) project by @io_Y_oi. It has ~42 issues and ~23 contributors.
6️⃣ Dolly is a large language model trained on the @databricks Machine Learning Platform, based on the GPT-J model.
It has been fine-tuned on a focused corpus of 50k records (Stanford Alpaca) to exhibit high-quality instruction-following behaviour. Dolly is intended exclusively for research purposes and is not licensed for commercial use 👀
The project has ~1 issue and ~4 contributors.
5️⃣ Next we havescrapeghost, a library for scraping websites using OpenAI's GPT API with the mission to make web scraping very easy. The repo includes a Python-based schema definition, HTML cleaning, CSS and XPath selectors, and auto-splitting.
The project has ~10 issues and ~1 contributor.
4️⃣ In number four we have xturing, a project that aims to provide an easy-to-use interface for fast and efficient fine-tuning of LLMs, such as GPT-J and GPT-2, with an aim to make it simple to personalise LLMs for your own data and applications.
It was brought to you by @stochasticai.
3️⃣ In the third place, we have BlenderGPT, an addon that enables users to use natural language commands to interact with Blender (a 3D modeling software). It's powered by OpenAI's GPT-3.5/GPT-4.
It has ~33 issues.
2️⃣ In second place we have Text2Video-Zero by @picsartai, which is the original repo for the Text2Video-Zero model. This model introduces a low-cost approach for zero-shot text-to-video generation without any training or optimisation.
The creators want to democratise AI and empower everyone's creativity, so they welcome external contributions. The repo has ~8 issues and ~6 contributors.
1️⃣ The top place goes to ZrrSkywalker/LLaMA-Adapter, which is the official implementation of the [LLaMA-Adapter], a new model that uses a small amount of data to turn a general language model into an instruction-following model. The method introduces a small number of learnable parameters and uses an attention mechanism to adaptively incorporate instructional signals, resulting in high-quality instruction-following sentences. A web demo is available.
The repo has ~6 issues and ~3 contributors.
We’ll write about trending repos on different topics and languages EVERY WEEK. Follow us here and on Twitter @quine to stay in the loop!
Do you want to build an impressive open source portfolio, but don't know where to start? We're here to help! Check out Quine and fast-forward your journey as an open source creator 🤗
Top comments (0)