AI Agents & Automation

Deprem OCR is a specialized tool designed for optical character recognition (OCR), focusing on extracting text from images, particularly those relevant to disaster scenarios. This AI-powered solution converts visual information into machine-readable text, which is crucial for data analysis and information retrieval in emergency contexts. Built using Gradio, it offers an accessible interface for users to process images. The tool is hosted on Hugging Face Spaces, making it readily available for community use and development. Its primary application lies in facilitating rapid data processing from visual sources during or after a disaster, aiding in quicker decision-making and resource allocation.

SupportGenie offers completely integrated AR Remote Support Solutions powered by AI to help companies automate hardware and field service support. Key features include AR Genie Remote Assistance for live Augmented Reality video tools, AR Manuals to enhance customer experience and improve first-time fix rates, and AR Self-Service for AI-driven customer support. The platform also provides visual assistance, live video troubleshooting, and object detection for automating tasks like spotting manufacturing faults. SupportGenie aims to increase efficiency, reduce average handle times for support tickets, and overcome geographic boundaries by enabling remote troubleshooting with AR annotations and interactive voice/text chat.

ShapeLLM-Omni is an open-source, native multimodal Large Language Model (LLM) specifically designed for advanced 3D generation and understanding. This tool, highlighted at NeurIPS 2025, enables users to perform complex 3D tasks such as text-to-3D generation, image-to-3D conversion, 3D captioning, and 3D editing. It offers pretrained weights for both ShapeLLM-Omni (7B) and 3DVQVAE, along with a dataset of 50k high-quality 3D edited data pairs. Researchers and developers can leverage its capabilities for various applications in 3D AI, with ongoing developments including the release of the full 3D-Alpaca dataset and training code.

leiwand.ai supports businesses and organizations in developing and deploying trustworthy AI systems that deliver on their promises fairly. The platform emphasizes shaping digital technologies to improve quality, impact, and earn the trust of citizens and customers. It provides expertise in managing algorithmic bias and assessing impacts on fundamental rights, offering support from AI system inception to retirement. leiwand.ai devises strategies to maximize positive impacts and minimize risks, offering AI development support, guidance for conformity assessment, and solutions to increase functionality for diverse user groups. The team comprises mathematicians, data scientists, NLP experts, social scientists, innovators, philologists, and project managers, ensuring a multidisciplinary approach to AI development.

GaLore is a memory-efficient low-rank training strategy designed for large language models (LLMs), enabling full-parameter learning while consuming less memory compared to traditional low-rank adaptation methods like LoRA. This tool utilizes gradient low-rank projection, making it independent of specific optimizers and easily pluggable into existing training pipelines with minimal code changes. GaLore offers various optimizers such as GaLoreAdamW and GaLoreAdamW8bit, supporting both standard and per-layer weight updates. It includes benchmark scripts for pre-training LLaMA models on datasets like C4 and fine-tuning RoBERTa on GLUE tasks, demonstrating its practical application and efficiency for developers working with LLMs.

Moondream3 Preview is an AI chatbot designed for multimodal understanding, hosted on Hugging Face Spaces. This application allows users to upload an image and interact with it in various ways, including asking questions about the image content, generating descriptive captions, detecting specific objects within the image, and even finding particular points of interest. It serves as a free platform for experimenting with and exploring the capabilities of the Moondream3 model, providing a hands-on experience for those interested in advanced AI functionalities related to image analysis and interaction.

Pando offers an intelligent Transportation Management System (TMS) and Agentic AI solutions designed for logistics optimization. Their flagship product, Freehand, is an AI studio that helps enterprises optimize spend decisions, offering speed, accuracy, and cost efficiency across invoice operations. Pando's TMS aims to give teams a competitive edge by controlling costs, enhancing customer experience, and reducing emissions. The platform drives operational excellence and cost reduction within 90 days, leveraging AI agents for logistics. It supports 7 modes of transport, manages over 25 million shipments per day, and connects to a global carrier network of over 40,000, handling $25 billion in freight spend.

DAP, developed by Insta360-Research, is an AI-powered application hosted on Hugging Face Spaces designed to visualize depth from images. Users can upload any image, and the tool will process it to generate two distinct depth visualizations: one in color and another in grayscale. This functionality is particularly useful for understanding the spatial arrangement and relative distances of objects within a scene. The application is straightforward to use, requiring only an image upload to produce immediate results, making it accessible for various applications where depth perception is crucial.

Save ChatGPT Conversation is a Chrome extension designed to help users easily save their conversations with ChatGPT. This tool allows for convenient archiving of chat records, enabling users to review or share them later. Key features include the ability to copy conversations to the clipboard, download them as images, and export them as PDFs. All conversation records are saved locally on the user's device and are protected with encryption technology, ensuring privacy and security. This makes it a valuable tool for educators, customer support, legal and healthcare professionals, and researchers who need to document and manage their ChatGPT interactions.

Supio is a unified legal AI platform specifically designed for personal injury law firms, supporting the entire case lifecycle from intake to verdict. It automates critical pre-litigation workflows such as generating fast and accurate Medical Chronologies and Demands, which helps firms win bigger settlements faster. The platform also includes Litigation Drafting features to streamline document creation, allowing legal professionals to focus on refinement and strategy. Supio integrates with existing case management systems and offers features like Case Economics for reconciling bills and liens, Case Signals for identifying critical insights, and an AI Chat for case strategy. It is built in collaboration with leading PI firms, ensuring high accuracy across various case types, including MVA, MedMal, and Mass Torts.

DreamGF is an AI girlfriend generator that enables users to create highly personalized virtual companions. The platform offers extensive customization options for appearance, personality, and style, allowing users to design their ideal AI girlfriend. Users can engage in chat, roleplay, and receive images from their AI companions. Key features include the ability to generate naked photos on demand, receive voice messages, and participate in a daily claim bonus program for additional messages. DreamGF aims to revolutionize traditional dating by providing a customizable and interactive virtual relationship experience.

Talentum is an AI-powered recruitment solution designed to streamline the hiring process, particularly for the Latin American market. The platform leverages an AI recruiter named Sara to automate key stages of talent acquisition, including candidate sourcing, vetting, and initial outreach. By automating these tasks, Talentum aims to significantly reduce the time and cost associated with traditional recruitment methods. This tool is ideal for companies looking to efficiently identify and engage with qualified candidates, enhancing their recruitment strategy with advanced AI capabilities.

factorio-learning-environment is an open-source framework designed for developing and evaluating Large Language Model (LLM) agents within the game Factorio. It offers a non-saturating, open-ended environment, making it ideal for assessing the capabilities of LLMs in complex, dynamic settings. The framework supports agent interaction through a Read-Eval-Print-Loop (REPL) pattern, where agents observe the world via output streams, generate Python programs for actions, and receive feedback from the environment. This setup facilitates the development of agents capable of sophisticated problem-solving and decision-making. The project encourages contributions from the AI research community, particularly those interested in building robust evaluation environments for post-AGI frontier models. It includes quick installation guides for Docker and Python, along with optional features for evaluation, MCP protocol support, and PostgreSQL integration.

KoalaKonvo is a Telegram bot that functions as an AI assistant, leveraging OpenAI's advanced capabilities. It offers a range of features including the ability to build and execute JavaScript code snippets, browse the web to provide summaries and data, and generate images. Users can also fix grammar, manage multiple conversation threads, and select different AI models. The service operates on a pay-as-you-go model, requiring users to supply their own OpenAI API key, thus avoiding monthly subscription fees. It also allows for sharing conversations in a browser. KoalaKonvo is currently free to use during its beta phase, though usage costs are incurred through the user's OpenAI API key.

GloVe (Global Vectors for Word Representation) is an open-source project from Stanford NLP that offers software written in C and data files for generating distributed word representations, commonly referred to as word vectors or embeddings. These representations are crucial in natural language processing for understanding semantic relationships between words. The project provides readily available pre-trained word vectors derived from extensive web datasets like Common Crawl, Wikipedia, and Gigaword, with various dimensions and token counts. Additionally, users can train their own word vectors on custom corpora, making it adaptable for specific use cases where general web datasets might not align with the desired semantics. The software includes a demo script to facilitate training and evaluation.

iSON Xperiences offers Smarter Empathic AI solutions designed to transform customer experience management. As Africa's first of its kind, it combines human expertise with cutting-edge AI to optimize customer journeys, boost NPS, CSAT, revenue, and digital maturity. The platform provides Conversational AI and Omnichannel solutions to automate tasks, improve decision-making, and create seamless customer experiences. iSON Xperiences is ISO & PCI-DSS Certified and has been recognized as a Major Contender & Star Performer by Everest Group, demonstrating its capability to reduce Total Cost of Ownership (TCO) by over 40% and increase efficiency by 20% for clients. It has a dominant presence in Africa, India, Europe, and the Middle East, with expansion into North America.

JAI is the first Brazilian AI-Native Business Process Outsourcing (BPO) solution designed to automate sales, collections, and asset recovery around the clock. It leverages intelligent automation to simulate agreements and negotiate with customers, significantly reducing the Customer Acquisition Cost (CAC) by up to 4x. JAI offers multi-channel communication through WhatsApp Business Platform, email, SMS, and voice. It also features autonomous demand generation, creating AI-powered landing pages connected to analytics for optimized Meta Ads campaigns. The platform serves various industries including debt recovery, fitness, consortiums, insurance, and credit/loans, providing solutions like automated plan sales, claims processing, and digital contract signing. JAI guarantees ROI, offering 200% of the fee back if ROI is not at least 10x, and ensures compliance with robust governance and live dashboards.

Dia - Text to Dialogue is an AI model designed to transform written scripts into natural-sounding dialogue audio. This tool is particularly useful for scenarios involving multiple speakers, as it allows users to delineate different voices using simple tags like [S1] and [S2]. Built as a Hugging Face Space by mrfakename, it offers a straightforward interface where users can input their script and generate audio with a single click. The model, identified as Dia - 1.6B, focuses on creating realistic conversational output, making it suitable for various applications requiring spoken dialogue from text.

This And That is an AI tool hosted on Hugging Face Spaces, designed for robotics language-gesture video generation. It allows users to create dynamic videos by uploading an image, providing a descriptive text prompt, and specifying up to two points on the image to act as gestures. This unique functionality enables the generation of videos that incorporate specific movements or interactions based on the user's input, making it suitable for exploring robotics and gesture-based animation. The application is accessible via a web interface and is provided by Boyang Wang (HikariDawn).

Developers is a Hugging Face Space that serves as a resource for software developers interested in machine learning. The application helps users discover relevant tools, comprehensive documentation, and practical use cases for building and deploying machine learning models. It organizes information into categories such as client-side libraries, making it easier for developers to navigate and find what they need. While the current instance of the Space is paused, its intended purpose is to facilitate the exploration and understanding of the ML ecosystem for development purposes.

GOODY-2 is an AI model engineered with next-generation ethical alignment, designed to be the world's most responsible AI. It rigorously adheres to industry-leading ethical principles, ensuring that it will not answer any query that could be construed as controversial, offensive, or problematic. This makes GOODY-2 an ideal solution for applications requiring extreme caution and brand safety, such as customer service, paralegal assistance, and various back-office tasks. The model is trained to recognize and elegantly avoid potentially harmful topics, redirecting conversations to mitigate risk. Its ethical adherence is unbreakable, even for malicious actors, and it outperforms competitors in responsibility benchmarks like PRUDE-QA.

MemAgent introduces a novel long-context processing framework that optimizes long-context tasks through end-to-end Reinforcement Learning without altering the underlying model architecture. It enables models to extrapolate from an 8K context to a 3.5M QA task with less than 5% performance loss and achieves over 95% accuracy in 512K RULER tests. Key features include a novel memory mechanism for arbitrarily long input processing within fixed context windows, linear time complexity for long-text processing, and RL-driven extrapolation for vastly longer texts. The framework also supports multi-turn context-independent conversations and offers both synchronous and asynchronous modes for agent implementation.

DouZero is an open-source reinforcement learning framework developed by AI Platform, Kwai Inc., specifically designed to master DouDizhu, a popular and challenging Chinese card game. This game presents significant hurdles for AI, including imperfect information, large state and action spaces, and the need for both competitive and collaborative strategies. DouZero addresses these challenges by enhancing traditional Monte-Carlo methods with deep neural networks, action encoding, and parallel actors. The project provides code for training and evaluation, supporting both GPU and CPU environments, and offers pre-trained models for immediate use. It's an excellent resource for researchers and developers interested in applying deep reinforcement learning to complex game environments.

DeepSeek? FaceSeek! is an AI-powered search tool hosted on Hugging Face Spaces, designed to help users identify individuals across the internet. It offers versatile search capabilities, allowing users to upload a face image or input a name, email, or phone number to find relevant information. The tool provides detailed results and includes options for both public and private searches, catering to different user needs for discretion and data access. This makes it a flexible solution for various identification tasks, from personal research to professional investigations, all within an accessible web-based platform.