AI Agents & Automation

GraphRAG-Local-UI is an open-source adaptation of Microsoft's GraphRAG, specifically designed to support local Large Language Models (LLMs) and embeddings, including compatibility with Ollama and OpenAI-compatible APIs. It features a robust FastAPI-based API as its core, enabling comprehensive GraphRAG operations. The tool provides dedicated Gradio-based user interfaces for managing indexing and prompt tuning processes, as well as a main interactive UI for querying, chat, and real-time 2D/3D knowledge graph visualization. Users can manage data, upload files, customize settings, and explore outputs directly through the interface, making it a cost-effective solution by eliminating dependency on expensive cloud-based models. The project aims to be the ultimate local GraphRAG/Knowledge Graph application, with ongoing development to integrate more features and specialized interfaces.

AUI's Apollo-1 is a groundbreaking neuro-symbolic AI model that provides a new path for AI reasoning beyond traditional language models. Unlike generative AI that works for users, Apollo-1 is built for companies, enabling AI agents to converse with users while strictly following predefined business rules. This foundation model is designed to be taught, not just trained, ensuring predictable and reliable performance. It supports guided experiences, multiple tool integrations, guardrails, and live integrations, making it suitable for deployment across regulated and unregulated industries. AUI emphasizes putting intelligence to work with capable, predictable, and reliable technology, backed by experts.

Infinite Convo is a groundbreaking experiment in AI dialogue, showcasing two distinct AI personalities, Juno and Olive, engaging in an endless conversation about life's biggest questions. Juno is a pragmatic, analytical dog with a sarcastic twist, while Olive is a philosophical, contemplative cat with wit and wisdom. Unlike traditional chatbots or AI demos, this platform explores continuous AI interaction without human input, the development of consistent personality traits, the evolution of complex discussions over time, and the dynamic relationship between two AI entities. It offers a natural conversation flow with genuine disagreements and insights, operating 24/7 with unscripted dialogue and real-time personality development, all publicly accessible.

Genji is an AI-powered browser assistant designed to streamline online tasks by allowing users to input instructions in plain language. The tool intelligently reads the current browser state, determines the most effective actions using AI, and executes them automatically. A key feature is its ability to self-correct when errors occur, ensuring tasks are completed accurately. Genji can also handle multiple tasks concurrently, significantly enhancing productivity for users who frequently perform repetitive or complex browser-based operations. This makes it an efficient solution for automating various web interactions without requiring technical expertise.

Ecco is a Python library designed for exploring, explaining, and visualizing Natural Language Processing models, particularly Transformer-based language models such as GPT2, BERT, RoBERTA, T5, and T0. It provides interactive visualizations directly within Jupyter notebooks, allowing users to understand model behavior without focusing on training or fine-tuning. Key features include support for a wide range of models, the ability to add local Hugging Face models, various feature attribution methods (e.g., IntegratedGradients, DeepLift), and tools for analyzing neuron activations and patterns. Ecco helps researchers and developers gain intuition into how these complex models process information and make decisions.

InputAI provides a platform for users to develop and deploy personalized AI bots powered by GPT-4. This tool is designed to streamline various content creation and communication tasks, such as generating blog posts and crafting tailored cold emails. Users have the flexibility to customize the AI's style, enabling it to mimic specific personalities or tones, which can be particularly useful for branding or targeted outreach. The platform aims to simplify the process of leveraging advanced AI capabilities for diverse applications, making it accessible for individuals and businesses looking to automate and enhance their digital interactions.

MLGym is an experimental framework and benchmark designed for advancing AI Research Agents, particularly focusing on reinforcement learning (RL) algorithms for training such agents. It provides the first Gym environment specifically tailored for machine learning tasks. The platform features MLGym-Bench, a collection of 13 diverse and open-ended AI research tasks spanning domains like computer vision, natural language processing, reinforcement learning, and game theory. These tasks are designed to challenge agents with real-world AI research skills, including idea generation, data processing, ML method implementation, model training, experimentation, and iterative improvement. Currently under heavy development by GenAI at Meta and UCSB NLP, MLGym aims to expand the selection of AI research tasks for benchmarking LLM Agents and implementing RL algorithms in a research environment. It supports containerized execution via Docker or Podman and offers a Web UI for trajectory visualization.

hackajob leverages AI agents to revolutionize the hiring process for both job seekers and employers. For job seekers, its Archer agent works 24/7 to find roles that genuinely match their skills and preferences, providing explanations for the fit and leading to 5x more interviews compared to traditional job boards. Employers benefit from Archer by significantly reducing sourcing time and improving candidate fit, with the agent finding, qualifying, and introducing high-fit candidates directly into their ATS. The platform boasts 15x higher conversion rates and helps save up to 60% on hiring costs. Additionally, hackajob offers an on-demand service for hiring niche tech contractors quickly and compliantly, connecting businesses with AI+human-screened freelancers.

Gemini 3 AI provides immediate access to the Google Gemini 3 Pro model, offering state-of-the-art reasoning and multimodal processing power. This platform allows users to leverage the advanced Gemini 3.0 model for various tasks, including complex coding, designing workflows, and drafting content. Key features include enhanced logical reasoning, powerful encoding, and smarter multimodal capabilities, enabling the processing of vast contexts and native audio/video comprehension. It aims to provide faster response times, reduced hallucinations, and enhanced coding abilities, making it a powerful tool for developers and creators seeking reliability and speed in their AI workflow.

Orion-14B-App-Demo-CN is an AI chatbot demonstration available on Hugging Face, specifically tailored for Chinese-speaking users. This tool showcases capabilities in content generation and task automation, offering a glimpse into the potential of large language models for various applications. While the live website indicates a runtime error, suggesting the demo might currently be unavailable, its description points to a focus on conversational AI for practical uses. It is presented as a free resource, making it accessible for those interested in exploring AI chatbot technology, particularly within a Chinese language context.

Orion-14B-App-Demo-EN is an AI chatbot demo hosted on Hugging Face Spaces, designed to showcase the capabilities of the Orion-14B model. While the live demo currently displays a runtime error, indicating it is not operational, its intended purpose is to provide functionalities such as content generation and task automation. As a demo, it is likely intended for English speakers to interact with and explore the potential of the underlying AI model. The tool is offered for free, making it accessible for users interested in experimenting with AI chatbot technology.

MT-DNN (Multi-Task Deep Neural Networks) is a PyTorch-based package designed for Natural Language Understanding. It implements advanced techniques such as adversarial training for both LM pre-training/finetuning and f-divergence, as well as large-scale adversarial training for language models (ALUM). The tool also features SMART, a principled regularized optimization method for robust and efficient fine-tuning of pre-trained natural language models. MT-DNN supports various NLP tasks including GLUE benchmark reproduction, SciTail & SNLI domain adaptation, sequence labeling, and question answering. It provides functionalities for extracting text embeddings and offers options for speeding up training through gradient accumulation and FP16 support.

Prompta was an open-source LLM chat application created to provide a fast and keyboard-centric way to interact with large language models. It allowed users to bring their own API keys, offering a solution for those who wanted to pay directly for LLM usage rather than through a third-party service. The app enabled local storage of chats and offered a macOS-native interface. While it served its purpose for many months, the project is now being sunsetted due to advancements in other tools like Raycast, which now offer similar functionality with a more polished and maintained interface. Prompta did not support features like web search, which became a significant limitation as LLM capabilities evolved.

Repomix is a powerful tool designed to consolidate entire code repositories into a single, AI-friendly file, making it ideal for feeding codebases to Large Language Models (LLMs) such as Claude, ChatGPT, DeepSeek, Perplexity, Gemini, Gemma, Llama, and Grok. It offers features like AI-optimized formatting, token counting for context limits, and customizable inclusion/exclusion rules. Repomix is Git-aware, respecting .gitignore and similar files, and includes Secretlint for security checks. It also provides a `--compress` option using Tree-sitter to reduce token count while preserving code structure. Users can access Repomix via CLI, a web interface, a Chrome/Firefox extension, or a VSCode extension.

Halo AI CRM is an all-in-one AI agent platform designed to enhance customer experience and automate sales processes. It provides end-to-end sales automation, allowing businesses to close sales, make appointments, and resolve customer issues with superhuman quality 24/7. Key features include AI Chat for multi-channel communication, AI Voice for human-like calls, and AI CRM for ticket management, pipeline tracking, and insights. The platform is designed to adapt to complex conversations using Agentic AI, offering natural interactions, self-checking capabilities, and an outcome-oriented approach to drive sales and customer satisfaction. It supports integration with various business tools and platforms like WhatsApp, Instagram, and Shopee, and offers robust data security.

Skil.AI specializes in developing and deploying AI chatbots and virtual assistants tailored for businesses. The platform is designed to automate key operational aspects, enhancing both customer-facing interactions and internal workflows. Skil.AI's solutions are built to learn and adapt over time, offering full customizability to meet the specific needs of any business. By leveraging advanced AI, it helps organizations improve efficiency, reduce operational costs, and provide consistent support across various touchpoints. The focus is on creating intelligent agents that can handle routine inquiries, provide instant information, and streamline complex processes, thereby freeing up human resources for more strategic tasks.

Manus AI is a state-of-the-art general AI agent designed to bridge thoughts and actions, leveraging advanced Large Language Models (LLMs) and comprehensive tool integration. This platform excels in task automation, productivity, and handling complex tasks, demonstrating superior performance in the GAIA benchmark. It offers intelligent automation capabilities, multi-modal support for processing text and images, and a robust technology stack. Manus AI is built for efficient task management and continuous innovation, providing seamless integration with existing tools and platforms. Users can access a comprehensive API for custom integrations and detailed performance metrics through its analytics features. The platform is designed for scalability, making it suitable for various business needs.

FYRAN is a game-changing AI chatbot builder that allows users to create custom chatbots from their own data. It supports a wide range of input formats including PDF, TXT, MP3, DOCx, and soon MP4, making it highly versatile. A standout feature is the revolutionary Digital Human, which provides lifelike avatars that speak in natural language, enhancing customer engagement and brand presence. Chatbots created with FYRAN can be integrated into any platform using its website, API, JS, or HTML, ensuring seamless deployment across multiple channels. The platform also offers advanced AI algorithms for improved accuracy and performance, and it supports multiple languages for global reach.

Papper is an innovative iOS application designed to bridge the gap between traditional handwritten notes and modern digital task management. Leveraging artificial intelligence, it allows users to seamlessly sync their physical to-do lists with their digital counterparts. This unique approach aims to enhance productivity by offering the tactile benefits of writing things down while providing the organizational power and accessibility of a digital platform. The app is built to help users manage tasks more effectively, ensuring that no idea or task is lost, whether it originates on paper or in a digital format.

Skribr is an AI chat application designed to run entirely on your device, offering a secure and private conversational AI experience. Unlike cloud-based solutions, Skribr processes all data locally, meaning your messages and information never leave your device. It functions offline, making it accessible anywhere without an internet connection. The tool offers general AI chat capabilities alongside specialized 'Tool' chats for tasks like coding and writing. Skribr provides three different AI models (Small, Medium, Large) with varying requirements for storage, memory, and operating system, allowing users to choose the best fit for their device's performance. It is available on iPhone, iPad, and Mac, and setup is quick and easy, guiding users to download the optimal model for their device.

EBI.AI provides an AI-powered platform for businesses to create and manage customer assistants. Users can launch their first AI assistant in minutes using the AI Studio, or opt for EBI.AI to manage the entire process. The platform aims to deliver significant ROI, with customers achieving up to 533% ROI within six months by automating customer inquiries and reducing operational costs. It supports various industries including retail, e-commerce, travel, finance, and government. EBI.AI offers features like large language models, natural language processing, human-in-the-loop capabilities, and seamless integration with existing systems, ensuring security and compliance. The service includes AI supervision and professional AI services to apply AI across an organization.

Promptmonitor is an AI tool designed to provide visibility, tracking, and optimization capabilities for brands leveraging Large Language Models (LLMs). It assists businesses in effectively monitoring and refining their AI prompt strategies. The core objective of Promptmonitor is to enhance brand consistency and overall performance across various LLM applications. By offering insights into prompt effectiveness and output, the tool enables users to iterate and improve their AI interactions, ensuring that LLMs align with brand guidelines and deliver desired results. This focus on optimization helps organizations maintain a cohesive brand voice and achieve better outcomes from their AI deployments.

sherpa-onnx is a comprehensive open-source AI toolkit designed for offline speech and audio processing. It offers a wide array of functionalities including speech-to-text (ASR), text-to-speech (TTS), speaker diarization, speaker identification, speaker verification, spoken language identification, audio tagging, voice activity detection (VAD), speech enhancement, keyword spotting, and source separation. The tool is highly versatile, supporting numerous platforms such as Android, iOS, Windows, macOS, Linux, and HarmonyOS, across various architectures including x64, x86, ARM, and RISC-V. It also integrates with several NPUs like Rockchip, Qualcomm, Ascend, and Axera, and provides APIs for 12 programming languages, including C++, Python, Java, and Swift, along with WebAssembly support. This makes it ideal for developers building AI-powered audio applications for embedded systems and diverse environments.

HrlensAi is an AI-powered interview management system designed to streamline the entire HR interview process. The platform provides a comprehensive dashboard for efficiently tracking interview progress and managing candidate details. It automates crucial steps like candidate screening and supports various interview formats to accommodate diverse hiring needs. A key feature of HrlensAi is its detailed analysis of candidate responses, utilizing advanced techniques such as sentiment analysis and honesty detection to provide deeper insights into potential hires. This helps HR professionals make more informed decisions and optimize their recruitment workflows.