Research & Education

ChatGLM 6B is an AI chatbot designed for conversational interactions, offering understanding in both English and Chinese. Users can type messages and receive responses from the assistant. The tool is built on Gradio, making it accessible for various applications. It is licensed under MIT, indicating its open-source nature and availability for free use. While the specific Space on Hugging Face is currently paused, the underlying model and its capabilities suggest its utility for research and general chatbot interactions, providing a versatile platform for conversational AI exploration.

ChatGLM2-SadTalker is an AI chatbot that combines conversational AI with voice cloning technology. This tool is primarily designed for research purposes and general chatbot interactions, allowing users to explore the integration of advanced language models with synthetic voice generation. It operates as a Hugging Face Space, making it accessible for experimentation and development within the AI community. The platform is built on Gradio, ensuring an interactive and user-friendly interface for testing its functionalities. Licensed under MIT, ChatGLM2-SadTalker is available for free, promoting open access and collaboration in the field of AI.

The Institute for Language and Speech Processing (ILSP) is a prominent research and development center dedicated to advancing language technologies. It engages in interdisciplinary research across linguistics and information technologies, focusing on areas such as language development, multilingual content processing, and speech technology. ILSP also plays a crucial role in developing its Language Resources Infrastructure, including projects like CLARIN:EL and initiatives for New Greek dialects. The institute offers various postgraduate programs in fields like Biomedical Informatics, Digital Humanities, and Language Technology, fostering education and innovation in these specialized domains.

MyEssayWriter.ai is an advanced AI essay writer designed to help students generate high-quality, plagiarism-free essays quickly. It produces structured essays with built-in APA, MLA, Chicago, and Harvard citations in under 90 seconds. The tool offers a free trial of up to 1500 words with no sign-up, credit card, or email required. It also includes features like AI detection resistance and an integrated AI Humanizer to refine output for strict AI policies. MyEssayWriter.ai supports various academic levels and essay types, making it suitable for high school, undergraduate, master's, and PhD students.

Chat with Baize is an AI chatbot developed by project-baize, available as a Hugging Face Space. It is designed to facilitate interactive conversations and allow users to explore language models. While the tool's specific functionalities are not detailed, its nature as a chatbot suggests capabilities for generating responses, engaging in dialogue, and potentially assisting with information retrieval or creative text generation. It is positioned for use in research and educational contexts, making it suitable for AI enthusiasts and researchers interested in experimenting with conversational AI. The tool is currently experiencing a runtime error, indicating it is not fully operational at this time.

Chat With CosmosLLaMa is an interactive AI chatbot powered by the Cosmos LLM, developed by the Yildiz Technical University Computer Engineering Department Cosmos Research Group. Users can ask any question and receive detailed responses by typing their message into the chatbox and pressing "Gönder." This tool is designed for general users, AI enthusiasts, and researchers interested in exploring the capabilities of large language models. It offers a straightforward interface for engaging in conversational AI, providing a platform for inquiry and information retrieval.

Eureka is an open-source research tool designed to bridge the gap in using large language models (LLMs) for complex low-level manipulation tasks in reinforcement learning. It leverages the zero-shot generation, code-writing, and in-context improvement capabilities of state-of-the-art LLMs like GPT-4 to perform evolutionary optimization over reward code. This allows Eureka to generate reward functions that surpass expert human-engineered rewards without task-specific prompting or predefined templates. The tool has demonstrated superior performance across 29 open-source RL environments, outperforming human experts on 83% of tasks with an average normalized improvement of 52%. Eureka also supports a gradient-free approach to reinforcement learning from human feedback (RLHF) and has enabled a simulated five-finger Shadow Hand to perform pen spinning tricks.

CogView4 is a Gradio demo of the CogView4-6B AI image generation model, designed to transform brief textual descriptions into elaborate prompts for creating diverse images. Users can specify whether they want realistic or stylized outputs, and fine-tune parameters such as image size and seed for greater control over the generated visuals. While the platform is currently paused, it offers a glimpse into advanced AI capabilities for image synthesis, making it suitable for research, development, and testing of image generation techniques. Its intuitive interface, built with Gradio, aims to simplify the process of AI-powered image creation.

Compressed Stable Diffusion is a Hugging Face Space application developed by Nota AI, designed for generating images from text prompts. This tool offers a unique opportunity to compare the output of a standard AI image generation model with a more efficient, compressed version. Users can input a text prompt and observe how each model interprets and visualizes the request, providing insights into the trade-offs between model size, efficiency, and image quality. It serves as an excellent resource for researchers, developers, and AI artists interested in the practical applications and performance differences of various AI models in image synthesis. The platform is web-based, making it easily accessible for experimentation and research.

Compare Llms is a Hugging Face Space designed for evaluating and contrasting the performance of various language models. Users can input a prompt and select from a range of available language models to generate text. The tool provides options to fine-tune the output by adjusting parameters such as temperature and maximum tokens, offering flexibility in controlling the generated content. This platform is particularly useful for educational purposes, research analysis, and anyone interested in understanding the nuances and capabilities of different AI chatbots. It offers a straightforward interface for direct comparison, making it accessible for both technical and non-technical users to experiment with AI text generation.

ControlNet-Video is an AI-powered tool designed for manipulating video content, offering capabilities for AI-assisted video editing and generation. The platform is built using Gradio, making it accessible for users to experiment with advanced video processing techniques. While the specific features for video manipulation are not detailed, the tool's foundation in ControlNet suggests it leverages conditional control for generating or modifying video frames based on various input conditions. This allows for precise adjustments and creative transformations within video projects. The tool is currently hosted on Hugging Face Spaces and is available for free, though it is currently paused.

Video Transcriber AI is an online tool designed to convert any video or audio file into accurate text transcripts in seconds. It supports a wide range of formats including YouTube links, Zoom recordings, MP4, MOV, and AVI files. Users can upload files up to 5GB and process up to 5 tasks simultaneously. The tool offers features like speaker recognition, multiple accuracy modes, and support for over 200 languages. It is free to use with no sign-up required, making it accessible for students, teachers, professionals, content creators, researchers, and journalists who need to quickly convert spoken content into editable text for various purposes.

Jada Ai is an augmented intelligence model focused on developing a general-purpose autonomous AI to drive business growth. The Jada Mark 0 prototype is engineered to be ethical and self-improving, adapting seamlessly to evolving operating necessities and environments. This advanced AI system integrates artificial intelligence, artificial general intelligence (AGI), and intelligent agents to accelerate business success. It is constructed based on the S3 Architecture, a multi-layer neural network, ensuring a robust and adaptable foundation for various business applications. Jada aims to help businesses join the AI economy and achieve significant growth through intelligent automation and adaptive capabilities.

Distil Whisper Web is an AI-powered speech-to-text transcription tool available as a Hugging Face Space. Users can upload audio recordings in formats like MP3 or WAV, and the application efficiently converts the spoken words into written text. The tool provides a clear and accurate transcription that can be easily copied, viewed, or downloaded for later use. It leverages Distil Whisper models, making it a valuable resource for anyone needing to transcribe audio files quickly and efficiently, from researchers to content creators.

Duke Applied Machine Learning (DAML) is an open-membership student organization at Duke University focused on advancing machine learning education. DAML provides comprehensive training pathways, including an eight-week Foundations Bootcamp for ML fundamentals and DevOps Workshops for deploying ML solutions. The organization also facilitates real-world experience through client projects, pairing Duke's CS talent with partners to design and deploy ML systems. Members gain practical skills in areas like data analysis, model prototyping, and documentation. DAML fosters an inclusive community, offering mentorship and opportunities to work on diverse projects, from AI chatbots to computer vision models for disease detection.

Document QA Bot is an AI-powered chatbot hosted on Hugging Face Spaces, designed to provide answers based on the content of uploaded documents. Users can easily upload a document, then select their preferred embedding model and Large Language Model (LLM) to process the information. This flexibility allows for customization in how the document's content is understood and queried. The tool then generates detailed answers to user questions, directly referencing the uploaded material. It's built with a Gradio interface, making it accessible for various users, and is licensed under MIT, indicating its suitability for research and educational purposes. The platform is free to use, making it an excellent resource for exploring document-based question answering.

E2/F5 TTS is an AI tool designed for zero-shot voice cloning, allowing users to generate audio from provided text using a reference audio clip. This unofficial demo, hosted on Hugging Face Spaces, offers two distinct Text-to-Speech (TTS) models for users to choose from. A key feature is its ability to transcribe the reference audio if no text input is given, providing flexibility in its application. The tool is built using Gradio, making it accessible for experimentation with advanced voice cloning technology. While currently experiencing a runtime error, its intended functionality focuses on creating synthesized speech that mimics a target voice.

Dreambooth Training is an AI tool hosted on Hugging Face that enables users to train custom Stable Diffusion models. Users can upload images of objects, people, or styles, and the application automatically adjusts the necessary settings for effective model training. Once trained, the custom model is uploaded directly to Hugging Face, making it accessible for further use or sharing. This tool simplifies the process of creating personalized AI models for image generation, catering to both technical and non-technical users interested in custom AI art or specific image manipulation tasks.

Echo-TTS Preview is a powerful text-to-speech (TTS) tool available on Hugging Face Spaces, designed for fast and efficient audio generation. It supports multi-speaker output and features advanced voice cloning capabilities, operating at a high fidelity of 44.1kHz. Users can input a text prompt and, for personalized results, provide a short voice recording to guide the output's voice. The application then creates a spoken-audio file, which can be saved in either WAV or MP3 format, closely matching the characteristics of the provided voice sample. This makes it ideal for creating custom audio content with consistent vocal styles.

F5 TTS Vietnamese 100h Demo is an AI-powered text-to-speech tool specifically designed for the Vietnamese language. It allows users to input text and provide a voice sample to generate natural-sounding Vietnamese speech. The application outputs both the generated audio and its corresponding spectrogram, offering a visual representation of the sound. This demo showcases a model that has been trained on an extensive dataset of 100 hours of Vietnamese speech, aiming for high-quality and authentic vocal output. The tool is built with Gradio, making it accessible as a web-based application.

Lindexer is an AI-powered systematic literature review software designed to streamline the entire workflow, from initial search to final reporting. Created by medical writers for medical writers, it aims to simplify the complex process of literature reviews, offering a user-friendly and highly customizable platform. Key features include direct search capabilities, smart AI-powered screening that sorts abstracts by relevance and suggests inclusion/exclusion decisions with justifications, and efficient data extraction with customizable forms and dual monitor support. The platform also provides built-in analysis and reporting tools, allowing users to generate datasets, calculate summaries, and export data to Excel. Lindexer emphasizes compliance, traceability, and team collaboration, with all pricing plans offering unlimited users.

Essay AI is an AI-powered writing assistant designed to streamline the essay writing process. It provides tools for grammar enhancement, ensuring polished and error-free submissions. Beyond basic writing support, Essay AI also incorporates plagiarism detection to help users maintain academic integrity and offers research assistance to aid in gathering relevant information. This tool is particularly beneficial for students, writers, educators, and researchers who need to produce high-quality written content efficiently and accurately. Its features aim to improve the overall quality and originality of written work.

HiRAG is an official repository for the paper "Retrieval-Augmented Generation with Hierarchical Knowledge," accepted to EMNLP 2025 Findings. This tool provides a robust framework for implementing and evaluating hierarchical knowledge in RAG systems. It allows users to index context, perform hierarchical searches, and integrate with various LLMs like DeepSeek, ChatGLM, and OpenAI. The repository includes examples for quick start, evaluation procedures, and comparative results against other RAG approaches, demonstrating substantial improvements in comprehensiveness, empowerment, and diversity of generated answers.

🔎NLP Image-Semantic-Search 🖼️ is an AI-powered tool hosted on Hugging Face that enables users to search for images using natural language processing. Users can input text queries to find related images, making it easy to locate specific visuals based on semantic meaning rather than just keywords. The tool offers advanced search capabilities, allowing for the combination of multiple queries using "; " and the exclusion of elements with "EXCLUDING ". This functionality provides a flexible and precise way to refine image searches, catering to various research and creative needs. It is freely accessible, making advanced image search technology available to a broad audience.