# llm

GraphRAG Visualizer is a web-based tool designed for visualizing and exploring data generated by Microsoft's GraphRAG framework. GraphRAG is a technology developed by Microsoft for generating graph structure data. The GraphRAG Visualizer allows users to easily view and analyze data by uploading Parquet files, without the need for additional software or scripts. Key features include graphical visualization, tabular data representation, search capabilities, and local data processing, ensuring data security and privacy.

A lightweight task engine designed based on principles of parallel execution. Ideal for developers and AI engineers to build complex workflows and AI agents. Free of charge.

AI-Data-Analysis-MultiAgent is an advanced AI-driven research assistant system that utilizes multiple specialized agents to assist with tasks such as data analysis, visualization, and report generation. The system employs LangChain, OpenAI’s GPT model, and LangGraph to handle complex research workflows while integrating a diverse array of AI architectures for optimal performance. A key feature of the system is its dedicated Note Taker agent, which maintains concise and comprehensive project records to reduce computational overhead, improve contextual continuity between different analysis phases, and deliver more coherent and consistent analytical outcomes.

Extractous is an unstructured data extraction tool written in Rust, offering multi-language bindings. It focuses on extracting content and metadata from various file types, such as PDF, Word, HTML, etc., with excellent performance and low memory usage. Extractous achieves fast processing speed and low memory consumption through native code execution, supports multiple file formats, and integrates Apache Tika and Tesseract-OCR technology for a wide range of file handling and OCR capabilities. The open-source nature and Apache 2.0 license allow for free commercial use, making it suitable for enterprises and developers handling large volumes of document data.

Sidecar is an AI plugin designed for the Aide editor, operating locally in conjunction with the editor. It is responsible for creating prompts, communicating with large language models (LLMs), and managing all interactions between them. Key advantages of Sidecar include improved programming efficiency, smart code completion, and integrated AI-assisted development. Developed in Rust, it ensures both performance and safety. Sidecar is suited for developers who require efficient programming and code management on local machines.

The pdf-extract-api is an API that utilizes modern OCR technology and Ollama-supported models to convert any document or image into structured JSON or Markdown text. It is built using FastAPI and employs Celery for asynchronous task processing, with Redis used for caching OCR results. The API has no reliance on cloud services or external dependencies, ensuring that all processing is completed in a local development or server environment, thereby safeguarding data security. It supports high-precision conversion from PDF to Markdown, including tabular data, numerical or mathematical formulas, and can convert PDFs to JSON using Ollama-supported models. Additionally, the API supports LLM-enhanced OCR results, removing personally identifiable information (PII) from PDFs, as well as distributed queue processing and caching.

rag-chatbot is an AI-based chatbot model that allows users to interact with multiple PDF files using natural language. This model leverages the latest machine learning technologies, such as Hugging Face and Ollama, to comprehend PDF content and generate responses. Its significance lies in its ability to process extensive document information, providing users with rapid and accurate question-answering services. The background information indicates that this is an open-source project aimed at enhancing document processing efficiency through technological innovation. Currently, the project is free and primarily targets developers and tech enthusiasts.

gptme is a personal AI assistant running in the terminal, equipped with local tools for writing code, using the terminal, browsing the web, visual recognition, and more. It serves as a local alternative to ChatGPT's 'code interpreter', free from limitations of software, internet access, timeouts, or privacy issues.

AI Artifacts is an open-source version of the Anthropic Claude Artifacts interface that utilizes E2B's Code Interpreter SDK and Core SDK to execute AI code. E2B provides a cloud sandbox for the secure execution of AI-generated code and supports tasks such as installing libraries, running shell commands, executing Python, JavaScript, R, and Next.js applications.

Composio is a platform that provides high-quality tools and integrations for AI agents, simplifying issues of authentication, accuracy, and reliability, allowing developers to integrate various tools and frameworks with just a single line of code. It supports over 100 tools and covers more than 90 platforms, including GitHub, Notion, and Linear. Composio offers diverse functionalities, including software operations, operating system interactions, browser features, search capabilities, software development environments (SWE), and real-time agent data (RAG). Additionally, it supports six different authentication protocols, significantly increasing the accuracy with which agents can call tools. Composio can also be embedded as a backend service in applications to manage authentication and integration for all users and agents, ensuring a consistent user experience.

AgentScope is an innovative multi-agent platform designed to empower developers to build multi-agent applications using large-scale models. It is easy to use, highly robust, and has distributed features based on the Actor model, supporting customizable fault tolerance control and retry mechanisms to enhance application stability.

Album AI is an experimental project that uses gpt-4o-mini as the visual model, automatically recognizes the metadata of image files in the album, and uses RAG technology to realize dialogue with the album. It can be used as a traditional album, or as an image knowledge base to assist large language models in content generation.

LazyLLM is a development tool that aims to simplify the process of building AI applications. It provides low-code solutions, enabling developers to easily assemble AI applications with multiple agents even without deep knowledge of large models. LazyLLM supports one-click deployment of all modules, cross-platform compatibility, automatic grid search parameter optimization, and efficient model fine-tuning, ultimately enhancing application performance.

Semantic Kernel is a software development kit (SDK) that integrates with large language models (LLMs) such as OpenAI, Azure OpenAI, and Hugging Face. It allows developers to interact with AI by defining chainable plugins, achieving AI integration within a few lines of code. Its key feature lies in the automatic orchestration of AI plugins, enabling users to generate plans for achieving specific goals using LLMs, which Semantic Kernel then executes.

OpenAgents is an open platform designed to enable users and developers to utilize and host language agents in their daily lives. The platform currently features three agents: a Data Agent for data analysis, processing, modification, and visualization; a Plugins Agent that integrates with 200+ everyday tools; and a Web Agent for automatic web browsing. OpenAgents empowers ordinary users to interact with agent functionalities through its optimized Web UI, while simultaneously providing developers and researchers with seamless local deployment experiences, laying the groundwork for innovative language agent construction and real-world evaluation.

June is a local voice chatbot that combines Ollama, Hugging Face Transformers, and Coqui TTS Toolkit. It provides a flexible and privacy-focused solution for voice-assisted interactions on a local machine, ensuring that no data is sent to external servers. Key advantages include offline accessibility, user privacy protection, and support for multiple interaction modes.

ChatTTS-Forge is a project built around the ChatTTS text-to-speech generation model. It provides a comprehensive API service and a Gradio-based WebUI, enabling the generation of long texts exceeding 1000 words while maintaining consistency. The platform boasts built-in style management with 32 distinct styles.

ModelScope-Agent is a customizable and extensible agent framework equipped with role-playing, large language model invocation, tool utilization, planning, and memory capabilities. It simplifies the implementation of agent applications by providing rich model and tool interfaces. Its unified interface and high extensibility, coupled with low coupling, allow developers to easily utilize built-in tools, LLMs, and memory components without being bound to higher-level agents.

AutoCoder is a novel model specifically designed for code generation tasks. It outperforms GPT-4 Turbo (as of April 2024) and GPT-4-o on the HumanEval benchmark dataset. Unlike previous open-source models, AutoCoder introduces a new feature: it can automatically install required packages and attempt to run the code when the user wishes to execute it, ensuring it works without issues.

llama-recipes is a companion repository for the Meta Llama model, aiming to provide an extensible library for fine-tuning the Meta Llama model and offering example scripts and notebooks to quickly get started with using the model in various use cases, including domain-specific fine-tuning and building LLM-based applications.

Farfalle is an open-source AI-powered search engine that allows users to run local large language models (LLMs) or utilize cloud models. It's based on the Perplexity clone and integrates with various technology stacks, including Next.js frontend, FastAPI backend, and Tavily search API. It also provides comprehensive documentation and demo videos for customized setups to help users get started quickly.

Create-tsi is a generative AI RAG (Retrieval-Augmented Generation) toolkit for low-code AI application generation. It leverages the large language models (LLMs) hosted by LlamaIndex and T-Systems on the Open Telekom Cloud to simplify the AI application creation process, making it quick and flexible. Users can use create-tsi to generate chatbots, write agents, and customize for specific use cases.

MaxKB is an enterprise-grade knowledge base question-answering system powered by LLMs. It aims to be the ultimate brain for businesses. Features include document upload, automatic online document crawling, and a smart question-answering experience. It also supports quick embedding into third-party business systems. The tech stack includes Vue.js, Python/Django, Langchain, and PostgreSQL/pgvector.

ctoc is a fast code statistic tool that can count the number of lines of code and tokens. It is an extension development based on the gocloc code statistic tool and is compatible with all features of gocloc. ctoc is particularly suitable for analyzing token levels in large code libraries and can aid in prompt engineering and model token cost estimation.