🧠 leverage advanced AI embeddings to perform multilingual zero-shot text classification. Whether you're dealing with unlabelled data or seeking to classify text against dynamic and user-defined labels, this library provides a seamless and efficient solution.

🚀 ClipServe: A fast API server for embedding text, images, and performing zero-shot classification using OpenAI’s CLIP model. Powered by FastAPI, Redis, and CUDA for lightning-fast, scalable AI applications. Transform texts and images into embeddings or classify images with custom labels—all through easy-to-use endpoints. 🌐📊

A Python toolkit and web API for detecting “bad” content (profanity, hate speech, gore, etc.) in text using a zero‑shot AI classifier.

Deep learning fish classifier combining ConvNeXt-Tiny (40 species, 98.96% accuracy) with BioCLIP-2 zero-shot recognition and AI-powered habitat mapping

AI Email Classifier is a full-stack smart inbox assistant that automates Gmail management using modern AI and real-time streaming. Incoming emails are pushed instantly to the backend via Google Pub/Sub webhooks, classified using Hugging Face zero-shot models, analyzed for sentiment/priority, and saved to a local SQLite database.

NoteFlow AI - Your smart companion for effortless document organization. Classify, summarize, and chat with your files, instantly. Transform your learning journey, one note at a time. This Modal is used to Classify the documents into desired classes using facebook/bart-large-mnli - Zero shot classification MNLI Model

🤖 AI-powered email automation tool using transformers & Gmail API. Automatically classifies emails (sales/support/finance/junk), applies smart labels, flags priorities & sends instant alerts for leads. Features zero-shot classification, continuous monitoring & customizable rules. Perfect for productivity!

No description available

AI-powered journaling classifier using zero-shot NLP

AI-powered multi-turn conversation intent classifier using BART zero-shot learning

AI-powered dress image classifier using OpenAI’s CLIP model for zero-shot fashion category prediction.

A Generative AI project to classify resumes using Hugging Face zero-shot classification

Zero-shot binary image classifier. Type in the name of an object and AI predicts whether your uploaded photo matches it.

Automatically classify and route HR tickets using AI. Built with n8n workflows, HuggingFace zero-shot classification, Slack notifications, and Google Sheets logging.

A smart movie recommendation system that suggests relevant movies to users by utilizing “all-MiniLM-L6-v2” (from the Sentence Transformers library) and a side AI component (BERT as a Zero-Shot Classifier).

Event-driven serverless architecture using zero-shot NLP classifiers with a rule-based inference engine to evaluate AI systems against multi-jurisdictional legal frameworks through document analysis.

A modular AI pipeline that classifies building types from Google Street View images using CLIP zero-shot vision and EasyOCR — no API key required, runs fully locally.

Auto Tagging Support Tickets Using LLM is an end-to-end AI project that automatically classifies free-text support tickets into top-3 categories using Large Language Models. It compares zero-shot and few-shot prompting approaches and demonstrates intelligent ticket triage automation.

Built an AI-driven Code Review Assistant that uses NLP to detect bugs, style issues, and security vulnerabilities in Python code. Integrated Flake8, Bandit, and a Hugging Face zero-shot classifier to explain code problems automatically. Implemented and demonstrated in Google Colab.

This project implements an AI-based Movie Genre Classification system that predicts the genre of a movie based on its plot summary. The classifier categorizes input text into one of three genres: Horror, Comedy, or Action, using a zero-shot learning approach.

A Streamlit app leveraging GPT-3.5-turbo for zero-shot sentiment analysis. Users input text and a list of potential emotions, the AI then classifies the text's sentiment and displays the corresponding emotion.

An AI tool that classifies news articles into topics using Hugging Face’s zero-shot classification pipeline. Provide the article text, and the model will return the most likely topics (e.g., politics, sports, technology, entertainment, business).

This project uses Generative AI to classify open-ended student feedback into ten categories such as Academics, Faculty, and Facilities. A synthetic dataset was created and analyzed using the FLAN-T5 model in a zero-shot setup, showcasing the potential of AI in automating educational feedback analysis.

SmartSupport-AI: An automated support ticket tagging system using Llama 3.3 (via Groq) to classify free-text queries into multi-class categories. Features Zero-Shot and Few-Shot learning for high-accuracy ranking of Top 3 tags. Includes a Gradio-powered web UI for real-time single ticket testing and bulk CSV batch processing

TalkToNews enhances news accessibility by extracting content from websites, reading it aloud with built-in TTS, classifying headlines using zero-shot AI models, and generating high-quality summaries using OpenRouter LLMs. Designed for quick navigation, improved accessibility, and faster comprehension of online news.

AI Question Difficulty Assigner is a hybrid NLP system that classifies questions into Easy, Medium, and Hard using rule-based heuristics and transformer-based zero-shot classification (BART-MNLI). It clusters students by performance and adaptively assigns questions with automated report generation.

Multi-label classification framework for mapping AI-related concerns from Twitter/X to UN Sustainable Development Goals (SDGs). Implements Zero-Shot, Fine-tuned BERT, and XLNet models to analyze 719,871 tweets (2022-2025). Achieved 96.4% accuracy with BERT in classifying public discourse on AI risks across 17 SDGs.

A lightweight, open-source AI system to **classify and tag social media posts** using image, text, and video inputs — and allow **open-vocabulary search** like Instagram's keyword search. Built with CLIP and Whisper, it supports flexible, scalable, and zero-shot classification and retrieval.

AI system that analyzes product reviews to detect sentiment, classify failure types, prioritize risk, and generate root-cause summaries using NLP. Built with Python, Pandas, Hugging Face Transformers, BERT sentiment analysis, zero-shot classification, and text summarization to create an automated early-warning intelligence tool.

AI-powered bird feeder camera running on Raspberry Pi 5 with Hailo-8 AI HAT+. Detects birds at ~22 FPS using YOLOv8, classifies species in real-time via BioCLIP zero-shot (42 Norwegian species), and streams live to the web with WebRTC. Includes web dashboard, species labeling UI, and full NUC/VPS relay pipeline.