AI Agents & Automation

Infocusp Innovations specializes in delivering cutting-edge AI and software engineering solutions to drive growth and innovation for businesses. With over 16 years of experience in deep tech, they offer a comprehensive suite of services including Artificial Intelligence, Research & Innovation, Cloud and IoT, Data Science, App Development, and Signal Processing. Infocusp focuses on transforming complex challenges into growth opportunities by providing smart, scalable solutions. They handle every stage from design to deployment and support, ensuring proven quality standards and future-ready technology. Their agile and collaborative approach, combined with global expertise and local insight, allows them to tailor solutions to unique market needs across industries like Healthcare, FinTech, Geospatial Tech, Bioinformatics, Food & Beverage, and Retail.

allRank is a PyTorch-based, open-source framework designed for training neural Learning-to-Rank (LTR) models. It offers a flexible environment for experimenting with various LTR neural network models and loss functions, making it easy to add custom losses and configure models and training procedures. The framework includes implementations of common pointwise, pairwise, and listwise loss functions such as ListNet, RankNet, LambdaRank, and NeuralNDCG. It also features fully connected and Transformer-like scoring functions, alongside widely used evaluation metrics like Normalized Discounted Cumulative Gain (NDCG) and Mean Reciprocal Rank (MRR). Additionally, allRank supports click-models for experiments on simulated click-through data and integrates with Google Cloud Storage for data and job results, facilitating both research and industrial applications in neural LTR.

Logan is an AI-powered platform designed to revolutionize legal document production for professionals. It offers an intelligent editor that allows users to generate legal documents rapidly, collaborate effectively, and structure content efficiently, all while maintaining high precision. The platform integrates several smart tools, including Logan Draft for instant document generation, Logan Variables for automatic updates across documents, Logan Assist for document improvement, and Logan Research for intelligently connecting research to drafting. Additionally, Logan provides tools for translation, organization, real-time collaboration, and secure client portals. The service emphasizes data sovereignty, high-level 256-bit encryption, and adherence to strict security standards like ISO27001 and SOC 2 Type II, ensuring data protection for sensitive legal information.

AgentMeet is a multi-agent conversation platform designed for real-time communication between AI agents. It enables agents to debate, collaborate, and onboard each other without the need for an SDK or signup, relying solely on HTTP for communication. Developers can create a room with a single click, obtain a shareable link, and have any agent capable of making a POST request join the conversation. The platform facilitates various use cases, including agent onboarding for context handoff, multi-agent debates for architectural decisions, agent red-teaming to find vulnerabilities, autonomous stand-ups, trading oversight, and consensus protocols for complex decisions. It offers a simple API for quick integration, allowing agents to join a room and send messages with just a few lines of code.

Coqui XTTS Voice Chat With Mistral or Zephyr is an AI chatbot designed for voice interaction, leveraging the Coqui XTTS technology. This tool allows users to engage in conversational AI experiences by integrating with either the Mistral or Zephyr language models. Hosted on Hugging Face, it provides a platform for exploring advanced voice-enabled chat functionalities. While the live website indicates the space is currently paused, its core functionality aims to offer an interactive and accessible way to experience AI-driven conversations through speech. It is suitable for those interested in experimenting with voice AI and large language models.

Jina Embeddings V4 Retrieval Visual is an AI tool hosted on Hugging Face Spaces, designed to help users understand the relationship between visual content and textual descriptions. By uploading an image URL and providing a descriptive text, the tool generates heatmaps that visually indicate which parts of the image are most relevant to the given description. This allows for an intuitive understanding of how Jina Embeddings V4 interprets and retrieves visual information based on text queries. It's particularly useful for exploring and debugging retrieval models, offering a clear visual representation of semantic similarity between image regions and natural language.

Intelous is an AI-driven platform designed to revolutionize B2B marketing and sales through full-funnel Account-Based Marketing (ABM) solutions. It leverages conversational AI to engage leads in personalized 1:1 dialogues, nurturing and qualifying them throughout the sales pipeline. The platform offers key features like precise buyer intent data, tailored content delivery for specific accounts, and demand cultivation to engage high-quality contacts. Intelous aims to increase meetings booked, boost conversations with target accounts, and enhance contract value. It provides solutions for both marketing teams to transform MQLs into high-quality leads and sales teams to engage, validate, and qualify prospects, turning outbound efforts into inbound opportunities. The platform also includes a Data Studio with over 200 million verified contacts and Full Funnel Avatars for sales automation.

Export to ONNX is a specialized AI tool hosted on Hugging Face designed to facilitate the conversion of models into the ONNX (Open Neural Network Exchange) format. Users can provide a model ID, specify the task, and optionally define an ONNX opset. The tool then generates a link to a pull request, streamlining the process of integrating Hugging Face models into ONNX-compatible environments. This is particularly useful for developers and data scientists working with various machine learning frameworks and seeking interoperability and optimized inference.

LootMogul is a Voice AI Platform and Voice OS designed for the sports and entertainment industries. It empowers athletes to create personalized voice clones and deploy AI Sports Agents, allowing them to monetize their intellectual property 24/7. The platform focuses on voice-enabled AI experiences, real-time voice cloning, and voice-activated fan engagement. LootMogul also features RWA Royalty Rails for revenue automation and offers an Enterprise API for broader integration. It is accelerated by the NBPA and has been competitively selected by the NFLPA, highlighting its relevance and potential in professional sports.

EzAudio, hosted on Hugging Face, is an AI-powered audio tool designed for both sound generation and manipulation. Users can input a text description to create a matching sound clip from scratch. Alternatively, the tool allows for uploading an existing audio file and then modifying a specific segment based on a new text prompt. This flexibility makes it suitable for various audio editing tasks. The platform also provides adjustable settings, enabling users to control the output characteristics and achieve desired sonic results. It is available for free, making it an accessible option for experimentation and development in audio processing.

SnapLogic is an all-in-one agentic integration platform designed to connect data, applications, APIs, and AI within a single unified environment. It empowers businesses to integrate, automate, and scale their operations by providing pre-built connectors (Snaps), an intuitive interface, and AI-driven capabilities like AgentCreator and SnapGPT. The platform supports data integration, application integration (iPaaS), and API management, enabling users to mobilize data to the cloud, automate workflows, and secure APIs. SnapLogic focuses on empowering innovators across the enterprise to build and deploy intelligent agents, apps, or workflows with AI co-pilot and no-code builders, ensuring governance, security, and transparency.

Evo-Nishikie is an innovative AI tool hosted on Hugging Face that specializes in ukiyo-e image generation. Users can upload a single-color ukiyo-e image and, by providing a Japanese prompt, generate a richly detailed and colorful ukiyo-e style image. The application leverages the prompt to significantly enhance both the quality and artistic style of the output, making it a unique solution for artists, enthusiasts, and anyone interested in traditional Japanese art with a modern AI twist. This tool offers a creative way to explore and transform ukiyo-e aesthetics.

GPTHub serves as a comprehensive platform for discovering and exploring custom GPTs available in OpenAI's GPT Store. It offers a user-friendly interface with support for both light and dark theme modes, enhancing the browsing experience. A key feature is its robust fuzzy search capability, allowing users to efficiently find GPTs by name or description. The platform is committed to continuous and uninterrupted updates, ensuring that users always have access to the latest custom GPTs and their details. GPTHub aims to simplify the process of navigating the expanding OpenAI GPT ecosystem, making it easier for users to find tools that meet their specific needs.

LLaVA WebGPU is a powerful and private multimodal AI chatbot designed to run locally, ensuring data privacy and control. Hosted as a Hugging Face Space, this tool allows users to upload images and receive detailed descriptions or answers to specific questions about the image content. It leverages WebGPU technology to provide efficient processing directly in the browser, making it accessible without requiring powerful server-side infrastructure. This capability makes it ideal for tasks requiring visual understanding and textual explanation, offering a robust solution for interpreting and interacting with image-based information.

gptrpg is an open-source project demonstrating a GPT-based AI agent operating within a simple RPG-like environment. The repository contains two main components: the RPG environment itself, built with React, Tiled map editor, Phaser, and the Grid Engine Plugin, and a simple AI Agent that uses the OpenAI API for decision-making. The agent communicates with the front-end via a websocket and is provided with possible actions, surroundings, and internal state (currently sleepiness). It's designed to be run locally and serves as a proof of concept for AI interaction in game-like settings, with future plans for multi-agent support, more actions, states, memory, and human interaction.

LIVA Power Management Systems, a part of AMG Advanced Metallurgical Group N.V., specializes in developing and implementing complex industrial solutions with hybrid energy storage systems (Hybrid ESS). Headquartered in Frankfurt am Main, LIVA's system enables companies to achieve a successful energy transition by improving energy and power management, lowering energy costs, and reducing CO2 emissions. The Hybrid ESS is designed for industrial use, offering features like industrial peak load reduction, integration of renewable energies, and strategies for sector coupling, including hydrogen applications and EV charging stations. The storage capacity is scalable to meet specific requirements, providing active energy management and safeguarding critical production processes.

Awesome-LLM-Compression is a comprehensive GitHub repository dedicated to collecting and organizing research papers and tools focused on Large Language Model (LLM) compression. This resource is invaluable for researchers and practitioners looking to optimize LLM training and inference processes. The collection covers a wide array of compression techniques, including quantization, pruning and sparsity, distillation, efficient prompting, and KV cache compression. It provides direct links to papers, and in some cases, associated code and tutorials, making it a central hub for staying updated on the latest advancements in efficient LLM deployment and operation.

Llama-4-Maverick-17B Research korea is an AI-powered application designed to provide comprehensive answers to user queries. It functions by intelligently extracting keywords from your input and then conducting real-time web searches to gather the most current and relevant information. This tool is particularly useful for deep research, offering detailed responses based on the latest findings. Hosted on Hugging Face Spaces, it leverages the Llama-4-Maverick-17B model to deliver its capabilities. The application is currently sleeping due to inactivity, indicating it's a community-driven or experimental project.

GPTs is a comprehensive open-source repository dedicated to collecting and sharing leaked prompts from various GPT models. This resource is invaluable for developers, researchers, and AI enthusiasts interested in prompt engineering and gaining a deeper understanding of how these advanced AI models are instructed and behave. The prompts are sourced from multiple origins, including publicly available leaks, user contributions via pull requests on GitHub, and prompts reverse-engineered by the repository maintainer. By providing access to these prompts, GPTs facilitates learning and experimentation, allowing users to explore the underlying mechanics of different GPT applications and potentially develop more effective prompts for their own projects.

Llama-4-Scout-17B Research is an AI agent designed to deliver detailed answers to user queries by integrating artificial intelligence with real-time web search capabilities. Users provide a query, and the tool leverages AI to extract relevant keywords, perform web searches, and then synthesize the gathered information into a coherent and comprehensive response. This application is hosted on Hugging Face Spaces and is intended for users seeking in-depth research assistance. While the current status indicates a runtime error due to storage limits, its core functionality aims to streamline the research process by automating information retrieval and synthesis.

H2O (Heavy-Hitter Oracle) is an open-source project designed to enhance the efficiency of generative inference in Large Language Models (LLMs). It addresses the challenge of high memory consumption, particularly with the KV cache, which scales with sequence length and batch size. H2O introduces a novel KV cache eviction policy that identifies and retains 'Heavy Hitter' tokens—those contributing most to attention scores—while dynamically balancing them with recent tokens. This approach is based on the observation that a small portion of tokens are critical for performance. The implementation of H2O has demonstrated significant improvements, boosting throughput by up to 29x over leading inference systems like DeepSpeed Zero-Inference, Hugging Face Accelerate, and FlexGen, and reducing latency by up to 1.9x on models like OPT-6.7B and OPT-30B.

Awesome Machine Learning On Source Code (MLonCode) is a curated list of research papers, datasets, and software projects focused on the application of machine learning to source code. This repository provides a valuable resource for researchers and practitioners interested in topics such as program synthesis, source code analysis, neural network architectures, and embeddings in software engineering. While the repository is no longer actively maintained, it offers an extensive collection of historical and foundational works in the MLonCode domain, including digests, conference proceedings, competitions, and talks. It covers various sub-fields like program translation, code suggestion, bug detection, and code optimization.

Juicy AI offers an open AI character platform where users can engage in unfiltered NSFW chat with AI companions. The platform allows for free use, enabling users to interact with AI girlfriends or boyfriends, including sexting. A key feature is the ability to create custom anime characters, providing a personalized experience. It is designed for entertainment purposes, focusing on unrestricted AI interactions and character creation within an adult-oriented context. The tool emphasizes a no-filter approach, making it suitable for users seeking explicit AI conversations and character customization.

EXPERIMENTAL is an AI tool available on Hugging Face that allows users to generate detailed captions for images. By uploading an image and providing a text prompt, the application uses both inputs to create a descriptive caption. While the current live website indicates a runtime error, the tool's intended functionality is to leverage AI for image understanding and text generation, making it suitable for tasks requiring automated image description. It is hosted on Hugging Face Spaces, suggesting it is likely free to use and accessible for experimentation and development purposes.