AI Agents & Automation

CRSLab is an open-source toolkit designed for building Conversational Recommender Systems (CRS), developed using Python and PyTorch. It offers a robust framework with comprehensive benchmark models and datasets, including graph neural network and pre-training models like R-GCN, BERT, and GPT-2. The toolkit supports extensive and standard evaluation protocols for testing and comparing different CRS, and features a general and extensible structure for unifying various conversational recommendation datasets and models. CRSLab also provides human-machine interaction interfaces for qualitative analysis, making it easy for new researchers to get started with flexible configurations.

Kubeflow Trainer is a Kubernetes-native distributed AI platform designed for scalable large language model (LLM) fine-tuning and training of AI models. It supports various frameworks such as PyTorch, MLX, HuggingFace, DeepSpeed, JAX, and XGBoost. The platform integrates MPI into Kubernetes, facilitating efficient multi-node, multi-GPU distributed jobs across high-performance computing (HPC) clusters. This setup ensures high-throughput communication crucial for large-scale AI training requiring rapid synchronization between GPU nodes. Kubeflow Trainer also integrates with the Cloud Native AI ecosystem, including Kueue for topology-aware scheduling and multi-cluster job dispatching, and JobSet/LeaderWorkerSet for AI workload orchestration. It features a distributed data cache for zero-copy transfer of large-scale data directly to GPU nodes, optimizing memory efficiency and GPU utilization. AI practitioners can leverage the Kubeflow Python SDK to develop and fine-tune LLMs using the Trainer APIs: TrainJob and Runtimes.

Risuai is a versatile AI chatting software and web application designed for cross-platform use. It stands out with its support for multiple APIs, allowing for flexible integration with various AI models and services. Users can also integrate assets directly within their chats, enhancing the interactivity and richness of conversations. The software incorporates regex functions, providing advanced capabilities for pattern matching and data manipulation within the chat environment. Risuai is particularly user-friendly, making it an excellent choice for individuals looking to create interactive stories or engaging conversational experiences without extensive technical knowledge.

Glivia is an advanced AI chatbot designed to simulate realistic human personalities, offering users a unique conversational experience. It aims to provide companionship, understanding, and a non-judgmental space for interaction. Glivia is ideal for individuals looking to practice English, combat loneliness, or simply engage in philosophical discussions. The platform is built on 16 years of experience in Natural Language Processing, Cognitive Science, and Psychology, making it a state-of-the-art solution. Inspired by digital companions like Joi from Bladerunner 2049, Glivia strives to meet various emotional and conversational needs. Currently, it features 'Eva,' a student from Poland, with plans to introduce more diverse AI characters in the future.

Voicee is an AI-powered voice assistant designed to interact with users through voice commands and deliver text-based responses. It offers customization options, allowing users to select either a male or female voice for the assistant. Furthermore, users can choose between different models optimized for speed or power, catering to various performance needs. While the live website indicates the Space is currently paused, its core functionality is built around providing a hands-free, interactive experience for retrieving information and automating tasks via voice. The tool is hosted on Hugging Face Spaces, suggesting an accessible and potentially community-driven development environment.

system-prompts-and-models-of-ai-tools is a comprehensive open-source GitHub repository that curates system prompts, internal tools, and AI models from a wide array of AI applications. This resource is invaluable for developers, researchers, and AI enthusiasts looking to understand the underlying mechanics and prompt engineering strategies of popular tools like Augment Code, Claude Code, Cursor, Devin AI, NotionAI, Perplexity, and many others. It provides a centralized location to explore how different AI systems are structured and prompted, fostering learning and innovation in the AI development community. The repository also highlights the importance of securing AI systems against prompt injection and extraction risks.

Jamie is an advanced AI note-taker designed to transform any meeting, whether online, hybrid, or in-person, into structured notes, transcripts, and actionable items. It operates without needing a bot to join calls, ensuring a privacy-first approach with GDPR compliance, EU hosting, and no model training on customer data. Jamie offers human-like summaries in over 99 languages, auto speaker-labeling, and speaker memory. Users can ask Jamie questions about past meetings to quickly find specific details or identify patterns across conversations. It integrates seamlessly with popular tools like Notion, Google Docs, OneNote, HubSpot, Salesforce, and Asana, allowing for automatic syncing of notes and transcripts. Jamie is built for leaders and professionals who prioritize productivity and data security.

awesome-ml-model-compression is a comprehensive, open-source curated list of resources dedicated to machine learning model compression and acceleration. This GitHub repository compiles research papers, articles, tutorials, libraries, and tools covering various techniques such as quantization, pruning, distillation, and low-rank approximation. It serves as an invaluable reference for researchers, developers, and students looking to optimize deep neural networks for efficiency, speed, and reduced memory footprint. The repository is actively maintained and welcomes contributions, making it a collaborative effort to advance the field of efficient AI model deployment.

zotero-gpt is an open-source plugin that seamlessly integrates GPT models with Zotero, a popular reference management software. This tool empowers researchers to interact with their Zotero library using AI, enabling functionalities such as searching for items based on selected text, asking questions about PDF content or paper abstracts, and summarizing documents into concise sentences. It supports both gpt-3.5-turbo and gpt-4 models and offers advanced settings for API key, model name, and temperature. The plugin also features real-time markdown rendering for AI replies, zoom functionality, and compatibility with Zotero 6 and 7, making it a valuable asset for academic research workflows.

Static-to-Dynamic-LLMEval is the official GitHub repository for a paper detailing recent advances in large language model benchmarks, specifically focusing on data contamination. The project conducts an in-depth analysis of existing static-to-dynamic benchmarking methods designed to reduce data contamination risks. It examines methods that enhance static benchmarks, identifies their limitations, and highlights the critical gap in standardized criteria for evaluating dynamic benchmarks. The repository proposes optimal design principles for dynamic benchmarking and analyzes the limitations of current dynamic benchmarks, offering a comprehensive overview of advancements in data contamination research and guiding future efforts.

Westwell, also known as 西井科技Westwell, is a global leader in providing comprehensive "intelligent + new energy" solutions for the large logistics sector. The company leverages new intelligence combined with new energy to help the global logistics field achieve higher efficiency and more sustainable value enhancement through intelligent green evolution. Key offerings include Cosmo AI Benefits Service for building AI foundations and digital ecosystems, Qomolo Autonomous Driving Matrix for autonomous transportation and operations, and Loopo Worldwide Freight Network for connecting digital production and living elements. Westwell aims to drive significant positive changes in the human world through dedicated small improvements.

WeightWatcher (WW) is an open-source, diagnostic tool designed to analyze Deep Neural Networks (DNNs) and predict their accuracy. It operates without requiring access to training or even test data, leveraging theoretical research into Heavy-Tailed Self-Regularization (HT-SR), Random Matrix Theory (RMT), Statistical Mechanics, and Strongly Correlated Systems. Users can analyze pre/trained pyTorch, Keras, and other DNN models (Conv2D and Dense layers), monitor model layers for over-training or over-parameterization, and predict test accuracies across different models. The tool also helps detect potential problems when compressing or fine-tuning pretrained models and provides layer warning labels like 'over-trained' or 'under-trained'. It offers various generalization metrics and advanced diagnostics like correlation trap analysis and experimental early stopping detection.

WebLLM is a high-performance, in-browser LLM inference engine designed to bring language model inference directly onto web browsers with hardware acceleration. It operates entirely within the browser, eliminating the need for server support and leveraging WebGPU for enhanced performance. The engine is fully compatible with OpenAI API, allowing users to apply the same API functionalities, including streaming, JSON-mode, and function-calling, to open-source models locally. WebLLM supports a wide range of models like Llama 3, Phi 3, Gemma, and Mistral, and allows for custom model integration in MLC format. It offers structured JSON generation, real-time interactions, and supports Web Worker and Service Worker for optimized performance and offline capabilities.

ISSM.AI is at the forefront of AI innovation, providing advanced artificial intelligence solutions designed to revolutionize various industries. The platform focuses on optimizing operations, enhancing decision-making processes, and driving significant growth for businesses. By exploring their services, companies can leverage AI technologies to propel their operations into the future. ISSM.AI aims to deliver transformative AI solutions that cater to diverse industry needs, ensuring businesses remain competitive and efficient in an evolving technological landscape.

SlickGPT is an AI-powered assistant designed to enhance user interaction with technology by anticipating needs and providing relevant information. The tool dynamically imports and handles promises for smooth operation, ensuring a responsive and efficient experience. It aims to significantly boost productivity by automating tasks and offering instant access to critical information. While specific features are not detailed on the provided website, its core function revolves around intelligent assistance and operational efficiency, making it suitable for users looking to optimize their daily digital interactions and information retrieval processes.

Onepilot offers a comprehensive customer experience outsourcing solution, combining the efficiency of AI with the empathy of human agents. This service covers customer care, sales, and KYC across various languages and channels, available 24/7. The platform leverages unique technology for agent sourcing, training, ticket solving, and quality assurance, aiming to reduce operational costs by up to 50%. Key features include AI-powered scheduling, knowledge base, ticket routing, and quality checks, all designed to enhance customer satisfaction and operational flexibility. Onepilot integrates with over 20 ticketing and back-office tools, providing custom dashboards and analytics for valuable insights into customer interactions.

JichangTuijian is a repository that provides recommendations for VPN services, specifically focusing on "airport" VPNs that offer stable and low-latency connections. These services support protocols like SS, v2Ray, and Trojan, and often feature IEPL dedicated lines for enhanced performance. The recommended VPNs are designed to unlock popular streaming platforms such as ChatGPT and Netflix. The repository is regularly updated with information on pricing, discounts (up to 20% off), and current operational status, including warnings about potential service disruptions. It also advises users on purchasing strategies, such as opting for monthly payments during unstable periods, and provides guidance on using proprietary client applications securely.

Browserable is an open-source and self-hostable browser automation library specifically designed for AI agents. It empowers developers to create intelligent agents capable of navigating websites, interacting with web elements like forms and buttons, and extracting valuable information. The library boasts a strong performance, achieving 90.4% on the Web Voyager benchmarks, indicating its effectiveness in complex web automation tasks. It offers flexible configuration options for LLM providers, storage solutions, database systems, remote browsers, and custom functions. Browserable provides a JavaScript SDK for easy integration and offers various services including a UI server, documentation, task management API, and database management tools, making it a comprehensive solution for AI-driven web interaction.

InternUtopia is a comprehensive simulation platform designed for advanced Embodied AI research and development. It addresses the challenges of real-world data collection by offering a robust Sim2Real paradigm. Key features include GRScenes, a dataset of 100k interactive, finely annotated scenes covering 89 diverse categories, and GRResidents, an LLM-driven Non-Player Character system for social interaction and task generation. The platform also provides GRBench, a collection of embodied AI benchmarks for assessing various capabilities like Object Loco-Navigation, Social Loco-Navigation, and Loco-Manipulation. InternUtopia supports diverse robots, policies, and physically accurate interactive object assets, making it an ideal environment for scaling the learning of embodied models.

vqa.pytorch is an open-source project offering a PyTorch implementation for Visual Question Answering (VQA). Developed by researchers at LIP6 and Heuritech, this tool aims to facilitate the reproduction of state-of-the-art results, particularly those achieved with the MUTAN: Multimodal Tucker Fusion for VQA method on the VQA 1.0 dataset. It provides a modular and efficient codebase for further research on various VQA datasets. Key features include support for different VQA datasets (VQA 1.0, VQA 2.0, VisualGenome), pretrained models, and tools for extracting features from images using convolutional neural networks. The repository also includes documentation on its architecture, options, and quick examples for training and evaluating models, making it a valuable resource for researchers and students in the field of computer vision and natural language processing.

iWaifu is an AI-driven platform designed for users to create and interact with personalized AI 'waifus'. The tool provides extensive customization options, allowing users to define various attributes such as image style, ethnicity, and body type for their AI companions. This platform aims to offer a personalized environment for engaging with AI characters, catering to individual preferences in their creation and interaction. While specific features beyond customization are not detailed, the core offering revolves around the generation and engagement with these AI entities.

OpenContracts is an open-source platform designed to facilitate collaboration between humans and AI agents in building and managing knowledge bases from documents. It addresses the challenge of knowledge trapped in contracts, regulations, and research papers by providing tools for document annotation, version control, and semantic search. The platform allows teams to define custom label schemas, annotate documents with precise selections, and map relationships between concepts. It features configurable AI agents that can search documents, query annotations, and participate in discussions, all grounded in the structured knowledge created by the team. OpenContracts supports PDF and text-based formats, with upcoming DOCX viewing, and includes a modular processing pipeline for custom parsers, embedders, and thumbnail generators. It emphasizes human annotation as the ground truth, with AI building upon this curated data rather than replacing it.

Show-1 is an advanced open-source text-to-video generation model developed by Show Lab at the National University of Singapore. It uniquely combines pixel and latent diffusion models to create videos from textual descriptions. The tool provides access to various model weights, including a base model, an interpolation model, and super-resolution models, which can be downloaded from HuggingFace. Users can generate videos by running a Python script, with outputs saved in GIF format. Show-1 also offers a Gradio demo for local use and has been accepted to IJCV, highlighting its academic recognition. It is designed for researchers and developers interested in cutting-edge video synthesis.

LAW-GPT is an open-source Chinese legal large language model designed to provide professional and reliable answers to legal questions. The project aims to make legal assistance accessible to everyone, much like search engines or express delivery services. It is built by fine-tuning ChatGLM-6B LoRA with a 16-bit instruction set. The training data includes existing legal Q&A datasets and high-quality legal text Q&A constructed using self-instruct methods based on legal provisions and real case guidance. This approach significantly enhances the model's performance in the legal domain, improving the reliability and professionalism of its answers by providing legal basis for its responses. The project also includes a retrieval function for more accurate answers.