In this research I'd like to use BERT with the huggingface PyTorch library to fine-tune a model which will perform best in question pairs classification. The app is build using Streamlit.

This repository contains a Streamlit app for retrieval-augmented generation on PDFs using HuggingFace models. Upload PDFs, ingest their content, and query information using natural language. The app supports GPU acceleration and multilingual queries with precise language detection. Ideal for efficient and accurate document analysis.

Simple app made with streamlit to analyze stock news and prices. Sentiment is given by a HuggingFace model, price prediction made with an XGBoost model.

This project applies classification models with the aim of automating the detection of toxic comments on social media. After choosing the model with the best performance, HuggingFace + Streamlit are applied to make the web app.

Some code I made to experiment with HuggingFace, SkLearn, Streamlit, and the Twitter API. The app scrapes Twitter for Market Sentiments using PySnscrape and processes them using a Roberta Natural Language Processing Model.

I developed an object detection model that can detect Vehicle, Pedestrians and Signboards with great accuracy by finetuning a yolo v8 on a curated dataset which I developed by by taking frames from a video and then by doing auto labeling with roboflow and then deployed the model on both huggingface space (Online) and streamlit App (Offline)

Deploy a Streamlit App on AWS EC2 - Cloud Computing Mini Project 9

This is a Gen-AI project with LangChain framework and Huggingface models for a research tool of equity and finance relate tools. The app is deployed in Streamlit where 3 news URLs are given as input to the LLM model and a query is given from user input. The LLM then returns results according to the input query.

DocumentGPT is a Streamlit-based app that allows users to upload documents and interact with their content using natural language queries. It processes uploaded files by chunking the text, generating embeddings with HuggingFace models, storing them in a Qdrant vector database, and utilizes Gemini 1.5 Flash to answer user queries based on Document.

asynchronous huggingface model app with streamlit

Streamlit app with HuggingFace transformers and OpenAI models for dessert suggestions

Streamlit based app with translation from huggingface.co model to translate from eng to de

ML App implemented with HuggingFace and Streamlit using a finetuned GPT-2 model trained with SpongeBob scripts.

End-to-end LLM-powered text summarization app with FastAPI backend, Streamlit frontend, and local HuggingFace model integration.

Multi-Task NLP Intelligence Suite A unified FastAPI + Streamlit app that runs 5 HuggingFace pipelines on any text input — with model comparison, confidence scores, and a live demo deployed to HuggingFace Spaces.

A Streamlit-based text classification web app using HuggingFace transformers. The model predicts emotions from user input text and displays the result in a clean UI. Built with Python, Streamlit, and a custom virtual environment.

The Language Model Analyst is a Python package and Streamlit app that enables natural language generation and analysis using HuggingFace-based language models (LLMs) and OpenAI GPT-3. This package and app are designed to simplify and streamline interactions with these powerful language models.

🚀 Chat with your PDF files using LangChain, FAISS, HuggingFace embeddings, and Groq’s LLaMA model - all wrapped in a Streamlit app. Upload PDFs, ask questions, and get professional AI-generated answers.

DBMind is an interactive Streamlit-based web app that lets users query any database using natural language. It uses LangChain with Google Gemini or HuggingFace models to generate, repair, and explain SQL queries intelligently.

A Streamlit app that summarizes news articles, analyzes sentiment using RoBERTa, and detects political bias with zero-shot classification. Supports text and image input (OCR). Built with transformers, torch, and HuggingFace models for real-time NLP insights.

A Streamlit app that lets you chat with your PDF documents. Upload a PDF, ask questions, and get accurate answers using open-source HuggingFace and Llama Index models. Perfect for quickly extracting information from your documents.

“A Streamlit app that combines Arxiv papers with LLMs using Retrieval-Augmented Generation (RAG). Search papers, get context-aware answers, and explore titles with expandable summaries. Powered by FAISS embeddings, HuggingFace models, and OpenRouter LLMs (Llama 4, Deepseek R1, GPT OSS).”

This is a Streamlit-based children's math learning app that uses LangChain with the HuggingFace moonshotai/Kimi-K2-Instruct model to solve basic mathematical problems interactively. It supports operations with single and two variables-- all powered by LLM tool execution.

A Streamlit-based emotion classification app that records or uploads audio, converts speech to text using Google Speech Recognition, and analyzes emotions with a HuggingFace DistilRoBERTa model. It displays dominant emotions and maintains transcription history for user interaction.

This project is a Streamlit-based RAG web app that lets users paste real-estate news URLs, builds embeddings with HuggingFace and ChromaDB, and then answers user questions using a Groq-hosted Llama 3 model grounded on those articles.

A simple Resume Question-Answering Chatbot built with Python, Streamlit, LangChain, SentenceTransformers, FAISS, and a HuggingFace FLAN-T5 model. Upload a PDF resume and ask questions; the app answers strictly from the resume using a lightweight RAG (Retrieval-Augmented Generation) pipeline.

WalletWatch is a simple AI-powered personal finance management app built with Python and Streamlit. Using IBM Watson, HuggingFace, and Granite models, it helps users track expenses, analyze spending habits, and get smart, location-based insights to improve their financial decisions.

This Streamlit app lets users upload and query multiple PDFs using NLP. It extracts, chunks, and embeds text with HuggingFace's Instruct model, storing it in a FAISS vector store for efficient retrieval. Powered by the `google/flan-t5-xxl` model, the app enables interactive Q&A based on document content, with a simple interface for seamless use.

This Streamlit app summarizes PDF, TXT, and MD documents using Retrieval-Augmented Generation (RAG) with Groq's LLaMA 3 model. It uses HuggingFace embeddings, FAISS for vector storage, and a powerful LLM pipeline to generate concise and accurate summaries from uploaded documents.

This Python-Streamlit app offers modular, multi-model AI selection with seamless switching. It processes user documents locally to provide context-aware LLM responses using APIs from Groq, Gemini, Perplexity, and HuggingFace. The UI is interactive, clean, and secures keys in session memory.