Welcome to the second edition of Hugging Fellows Highlights! In this page, we will go through what our Fellows accomplished past couple of weeks.
<aside> 🤗 You can learn more about Fellows program here.
</aside>
🧑🎨 Pi has released 22h diffusion v0-2, with support for different aspect ratios. Check out the model repository here and see the demos built on top of it 👉 https://huggingface.co/22h/vintedois-diffusion-v0-2
🧑🎤 Thomas Chaigneau has open-sourced an automatic speech recognition API built on faster-whisper and pyannote: https://github.com/Wordcab/wordcab-transcribe
He also fine-tuned LLaMa-7b and LLaMa-13b on the natural-instructions dataset using PEFT 🌟
🦹 Simon Dürr has built a Gradio App to create pull requests on GitHub automatically https://bioicons.com/submit/
👨🎨 Johannes Kolbe has built cool apps for Keras DreamBooth sprint. He fine-tuned Stable Diffusion on Return to Monkey Island Art Style and Marvin the Paranoid Android.
👨🔬 Aritra has created this tutorial on temporal latent bottleneck.
👩🎤👨🎤 María and Manuel have organized the second edition of NLP in Spanish hackathon, with +500 participants to the workshops and the hackathon 🤩🚀 Learn more about the Somos NLP Hackathon here. They have also organized a community challenge to clean the Spanish translation of ALPACA dataset, see the cleaned version here.
👨🎤 Manu released BLOOM fine-tuned on ALPACA dataset, Alpacoom. BLOOM 7B1 fine-tuned on Alpacas dataset using LoRA with PEFT. 🌸
He also started releasing BART-like models for Spanish in various domains, see them below.
🦸♀️ Shivalika has won a hackathon on multilingual semantic search and open-sourced her solution. You can find the Space and the GitHub repository. 🏅
👨🔬 Sylvain has released the code for simple cel animation engine, find the Space here. He has also added a ControlNet layer to generate video from cel-animation, find the Space here
He also released following Spaces that convert video to various feature extracted versions👇