AI Agents & Automation

deep-learning-with-keras-notebooks is an open-source collection of Jupyter notebooks designed to help users learn and apply Keras for deep learning. This repository provides a wide range of examples, from image processing and augmentation to advanced topics like object detection with YOLOv2 and natural language processing with word embeddings. The notebooks cover practical applications such as image classification (e.g., traffic signs, fashion MNIST), facial recognition, and captcha breaking. It's an excellent resource for students and developers looking to gain hands-on experience with Keras and deep learning concepts, offering clear, runnable examples for various tasks.

Firefly is a comprehensive open-source project designed for training large language models, offering support for pre-training, instruction fine-tuning, and DPO (Direct Preference Optimization). It is compatible with numerous mainstream models such as Qwen2.5, Qwen2, Yi1.5, Phi-3, Llama3, Gemma, MiniCPM, and many others. The platform facilitates full parameter training, as well as efficient training methods like LoRA and QLoRA, making it accessible even with limited computational resources. Firefly also integrates with Unsloth for accelerated training and reduced VRAM usage, and provides curated instruction fine-tuning datasets. It offers pre-trained Firefly series models and has demonstrated effectiveness on the Open LLM Leaderboard.

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.

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.

Hephaestus is a semi-structured agentic framework designed for building dynamic AI workflows. Unlike traditional frameworks that require predefined instructions for every scenario, Hephaestus allows agents to discover and create tasks based on their findings. It defines logical phase types (e.g., Plan, Implement, Test) and lets agents spawn new tasks in any phase, leading to self-branching workflows. This approach ensures adaptability, as the workflow evolves based on actual discoveries rather than anticipated scenarios. It includes features like real-time monitoring, Kanban board coordination, and dependency tracking, making it ideal for complex software development projects where agents can identify optimizations or bugs and create new work to address them.

KS Solutions AG is a provider of industrial software and AI solutions, focusing on enhancing efficiency and quality in industrial processes. They offer specialized AI tools and interfaces tailored for industrial applications, including software development, AI integration, and adherence to industry standards. Their services encompass consulting and the development of custom AI solutions, such as their products SyntaxShift and DynoCode. The company emphasizes improving automation, code generation, and quality assurance within industrial settings, making them a partner for businesses seeking advanced technological solutions in this domain.

langchain-kr offers a comprehensive Korean tutorial for LangChain, built upon the official LangChain documentation, cookbooks, and practical examples. This resource is designed to help Korean speakers understand and utilize LangChain with greater ease and effectiveness. The tutorial covers a wide range of topics, from basic concepts and prompt engineering to advanced techniques like RAG, LangChain Expression Language (LCEL), and multi-agent collaboration with LangGraph. It includes practical examples, YouTube video explanations, and blog posts, making it a valuable learning resource for anyone looking to master LangChain in Korean. The project is open-source and encourages contributions from the community.

SermoAI is an AI chatbot specifically developed for the banking and fintech sectors, aiming to revolutionize customer support. It offers multilingual capabilities, enabling financial institutions to serve a diverse global customer base efficiently. The tool is designed to automate customer interactions, providing instant assistance and reducing response times, which significantly enhances customer satisfaction. By integrating SermoAI, businesses can improve operational efficiency, handle high volumes of inquiries, and ensure consistent, high-quality customer service around the clock. This specialized AI solution helps financial companies streamline their support operations and maintain a competitive edge in a fast-paced industry.

Latimer is a large language model designed to provide a more accurate and inclusive understanding of diverse histories and cultures. Unlike traditional AI models that may exhibit biases due to their training data, Latimer is specifically trained with a focus on diverse historical narratives and a culturally fluent voice. This approach allows it to better reflect the experiences, cultures, and histories of various communities, aiming to offer a more balanced and representative perspective. It is built to serve as a valuable resource for individuals and organizations seeking information that is sensitive to cultural nuances and historical accuracy, promoting a more inclusive AI experience.

MacBERT is a sophisticated pre-trained language model specifically designed for Chinese Natural Language Processing (NLP). It builds upon the foundational BERT architecture by incorporating a novel Masked and Corrected (Mac) language model pre-training task. This innovative approach aims to mitigate the common 'pre-training-downstream task' inconsistency, a challenge where the [MASK] token used during pre-training is absent in real-world downstream applications. MacBERT addresses this by replacing [MASK] tokens with similar words, derived using a synonyms toolkit based on word2vec similarity. It also integrates Whole Word Masking (WWM) and N-gram masking techniques. The model maintains full compatibility with BERT, allowing for seamless integration into existing NLP workflows without code modification. MacBERT has demonstrated significant performance enhancements across various Chinese NLP tasks, including extractive question answering, natural language inference, sentiment classification, and sentence pair matching.

go-proxy-bingai is an open-source demonstration site for Microsoft's New Bing, built with Vue3 and Go. It aims to provide a consistent user interface experience similar to the official New Bing, while also supporting ChatGPT prompts. This tool is particularly useful for users in regions where access to the official Bing AI might be restricted, offering a viable alternative. It allows for local deployment and can be configured with various environment variables for customization, including proxy settings and user cookies for advanced features like image creation. The project emphasizes ease of deployment with options like Docker, Railway, Vercel, and Render, and includes a separate chat server deployment option for enhanced stability and availability.

gpt-assistant-android is an open-source, full-featured GPT assistant designed for Android devices. It offers convenient activation via volume keys for voice interaction, enabling seamless communication with the AI. Key capabilities include internet access for real-time information, photo capture, and comprehensive document parsing for formats like TXT, PDF, DOCX, PPTX, and XLSX. The tool also features intelligent templates for customized interfaces, multiple voice input/output options, and an experimental agent mode that allows the AI to control phone functions like clicking and scrolling. Users can configure their own OpenAI API keys or use third-party forwarding services, making it a versatile and powerful personal assistant for Android users.

LLM-RL-Visualized offers a comprehensive collection of over 100 original architectural diagrams to systematically explain large language models (LLMs) and reinforcement learning (RL). This resource delves into the core principles of LLMs and Vision-Language Models (VLMs), various training algorithms such as RLHF, GRPO, DPO, SFT, and CoT distillation, as well as optimization techniques like RAG. Authored by the creator of "Large Model Algorithms," it serves as a valuable visual aid for understanding complex AI concepts. The repository is continuously updated with corrections and additions, providing high-definition diagrams and scalable SVG vector images for detailed study.

PROTAC Scientific-Drug Discovery Pro provides comprehensive computational services to accelerate drug lead discovery. The platform utilizes cutting-edge machine learning and informatics tools, alongside in-house multidisciplinary expertise, to streamline the drug discovery process from strategy development through publication. Services include fragment-based lead discovery, molecular docking, protein homology modeling, pharmacophore modeling, and the creation of customized machine learning models for bioactivity prediction. They also offer training programs and workshops, and support for publishing in top-tier journals and patenting. The company aims to reduce costs and save time for researchers, with financial support options for students and referral programs.

Polyglot is a versatile, cross-platform desktop application, also available as a web version, designed for AI-powered language practice. Leveraging ChatGPT and Azure AI language models, it provides an intuitive platform for users to engage in multilingual oral practice. Key features include intelligent speech synthesis, smart conversation capabilities, and support for multiple languages with built-in AI characters from different countries. Users can customize languages, AI personas, and configure their own Azure and OpenAI API keys. It also offers dark mode, text translation, custom avatars, and voice recognition with hotkey support, making it a comprehensive tool for language learners.

Lime is a desktop application built with Tauri, designed for creators, content teams, and knowledge workers. It consolidates Workspace, Agent, Skills, MCP, Claw channels, and Artifact delivery into a unified environment. This allows users to transform initial ideas into publishable content, including drafts, images, and videos. The platform supports the entire creative process from ideation to final delivery, enabling content generation, image production, video collaboration, and project asset accumulation. Key features include a Workspace for organizing tasks and assets, Skills for defining interactions and workflows, MCP as a standardized capability layer, and Claw channels for asynchronous collaboration through platforms like Feishu and Telegram.

stable-diffusion-webui-extensions serves as the official extension index for the Stable Diffusion Web UI, providing a centralized repository for users to find and integrate additional functionalities. This open-source project allows developers to submit their extensions, which are then reviewed and added to the index, making them accessible to a wider user base. The platform facilitates the customization and enhancement of stable diffusion workflows by offering a variety of extensions, each tagged for appropriate categorization. It includes guidelines for submitting new extensions, ensuring they are functional and properly described. The index also provides important tags like 'online' for extensions requiring external server connections and 'ads' for those containing advertisements, promoting transparency for users.

VnCoreNLP is a comprehensive Vietnamese natural language processing toolkit, designed to provide fast and accurate linguistic annotations. It integrates essential NLP components such as word segmentation, part-of-speech (POS) tagging, named entity recognition (NER), and dependency parsing. Users can run processing pipelines either from the command-line or through its API, eliminating the need for external dependencies. The toolkit supports both Python (via a wrapper) and Java, making it accessible to a broad range of developers and researchers. VnCoreNLP's architecture and experimental results have been published in prominent NLP conferences, highlighting its effectiveness for Vietnamese text analysis.

Warp-yg is a comprehensive, multi-functional script designed for managing WARP configurations. It offers seamless switching between warp-go and wgcf, providing flexibility for users. A key feature is its ability to generate an unlimited number of WARP-Wireguard configuration files, catering to diverse needs. The tool also supports upgrading WARP+ and WARP team accounts, enhancing connectivity options. Beyond configuration, Warp-yg allows users to check their VPS local IP address and determine the Netflix and ChatGPT unlock status, which is crucial for users relying on these services. The script is compatible with pure IPv4 and IPv6 VPS installations and supports mainstream Linux systems, making it a versatile solution for network management.

Whisper-Finetune is an open-source project designed to fine-tune the Whisper speech recognition model. It offers flexible training options, including support for data with or without timestamps, and even training without speech data. The tool significantly accelerates inference processes and provides versatile deployment capabilities across Web, Windows desktop, and Android platforms. It leverages techniques like Lora for fine-tuning and supports CTranslate2 and GGML for accelerated inference. The project includes detailed instructions for environment setup, data preparation, single and multi-GPU training, model merging, evaluation, and various prediction interfaces, making it a comprehensive solution for customizing and deploying Whisper models.

TransGPT is the first open-source large language model specifically designed for the transportation industry in China. It aims to provide practical value by offering functionalities such as traffic condition prediction, intelligent consultation, public transportation services, traffic planning and design, traffic safety education, management assistance, and accident reporting and analysis. The model also supports autonomous driving assistance systems. TransGPT serves as a general knowledge base for various transportation sectors, including road, bridge, tunnel engineering, highway and waterway transportation, and urban public transport. It is available in two main models: TransGPT-7B and TransGPT-MM-6B, with both text and multimodal capabilities. The project provides training and inference code, along with commercial-use-approved datasets for pre-training and fine-tuning.

Tinyflow is a lightweight, open-source AI agent solution designed as a development component rather than a standalone product. It enables developers to integrate AI agent orchestration capabilities into existing applications. The frontend is built with Web Components, ensuring compatibility with popular frameworks like React, Vue, Angular, and Svelte, as well as native HTML, CSS, and JavaScript. For the backend, Tinyflow supports various languages including Java, Python, and Node.js, with Java backend implementation available and Python/Node.js versions currently under development. This flexibility makes Tinyflow a versatile tool for enhancing traditional applications with advanced AI agent functionalities.

GPTKeeper is a Chrome extension designed to simplify the process of saving and sharing ChatGPT conversations. This tool allows users to easily store their interactions with ChatGPT, making them readily accessible for future reference, documentation, or sharing with others. It's particularly useful for documenting research, facilitating project collaboration by sharing relevant AI interactions, or maintaining a personal archive of interesting and important conversations. The extension aims to streamline the management of AI-generated content, ensuring that valuable insights and discussions are not lost and can be easily retrieved or distributed.

Mockaroni is an AI-powered platform specifically developed to help job seekers practice and improve their interview skills. The tool simulates real-world interview scenarios, allowing users to gain experience and confidence before actual job interviews. It offers personalized feedback on performance, helping users identify areas for improvement in their responses, body language, and overall presentation. Mockaroni aims to enhance a user's ability to articulate their skills and experiences effectively, making it a valuable resource for anyone looking to boost their chances of success in the competitive job market.