Projects

### Overview: This is an unofficial Go (Golang) bindings for the OpenAI API, providing comprehensive support for various functionalities including audio, chat, completions, embeddings, files, fine-tuning, images, models, and moderations. ### Key Features: - Audio - Chat - Completions - Embeddings - Files - Fine-Tuning - Images - Models - Moderations ### Tech Stack: Go ### Why it Matters: This library is essential for developers looking to integrate OpenAI's capabilities into their Go projects, offering a straightforward and efficient way to interact with the API. It is maintained by the community and is not affiliated with OpenAI, making it a valuable resource for those seeking to leverage the API without official support.

Overview: This project provides a comprehensive set of 15 RAG (Reinforcement Learning for Answer Generation) modules, including document loaders, guardrails, and a research paper assistant with citation and claim verification capabilities. Additionally, it includes LangGraph workflows and LangSmith tracing for advanced AI development. Key features: - 15 RAG modules - Document loaders - Guardrails - Research paper assistant with citations and claim verification - LangGraph workflows - LangSmith tracing Tech stack: Jupyter Notebook Why it matters: This project offers a practical toolkit for developers looking to enhance their AI development skills, particularly in the areas of research paper assistance and advanced workflow management.

Overview: Entity Neutering is a methodology designed to pre-process text data to prevent lookahead bias in Large Language Models (LLMs), particularly in financial text analysis. It anonymizes financial documents and news articles by masking and/or paraphrasing identifying features, ensuring that model predictions are based solely on the textual content. Key features: - Iterative masking and paraphrasing until the same LLM cannot infer the entity. - Optional LLM-based masking and paraphrasing. - Flexible processing modes: masking only, paraphrasing only, or complete neutering. - Up to 8 rounds of refinement. - Built-in identification testing to verify neutering success. - Robustness testing with Ollama models. Tech stack: Python. Why it matters: This methodology helps ensure that LLMs do not use prior knowledge about specific companies, industries, or time periods, thereby preventing lookahead bias and improving the fairness and reliability of financial text analysis.

### Overview: FlexInfer is a mobile-oriented LLM inference runtime designed to run large quantized models under tight memory budgets. It extends the `llama.cpp` and `ggml` stack with prefetch-aware execution paths, memory planning, and Linux host plus Android experiment scripts. ### Key features: - Prefetch-aware execution paths - Memory planning - Linux host plus Android experiment scripts ### Tech stack: C++ ### Why it matters: FlexInfer provides a flexible and efficient way to run large quantized models on mobile devices, optimizing memory usage and performance. It enables developers to leverage powerful LLMs without the constraints of high memory budgets, making it suitable for a wide range of applications.

Overview: omlx is an application designed to optimize large language model (LLM) inference on Apple Silicon Macs, providing efficient performance through continuous batching, SSD caching, and seamless menu bar control. Key features: - Continuous batching - SSD caching - Menu bar control - OpenAI API compatibility - Lightweight system requirements Tech stack: Python Why it matters: omlx simplifies the use of LLMs on Macs, making large language model inference faster and more efficient, while managing resources effectively to ensure smooth performance. Repository: Mizistein/omlx Primary language: Python GitHub stars: 8 GitHub description: 🤖 Optimize LLM inference on Mac with continuous batching and SSD caching managed from your menu bar for efficient performance.

### Overview: Irminsul is a Genshin Impact AI assistant that answers questions based on game data, refusing AI-generated content and leaks. It uses a combination of QLoRA, Pinecone, and FastAPI to serve answers in multiple languages, ensuring source-cited responses and a leak-gated approach. ### Key Features: - **Answers from Game Data:** Every answer links to its source, and game data is marked as *No AI used*. - **Leak-Gated:** Questions that cannot be proven released before indexing are refused. - **Builds and Teams:** Answers are checked against game data, with newer characters using quick guides or stating that no build exists yet. - **Quests and Lore:** Answers are dated to prevent story spoilers. - **Live Game News:** Banner updates, event information, and expired codes from HoYoverse. - **Spoiler-Free Lore:** Answers stop at the player's Archon Quest progress. - **Supports Multiple Languages:** Answers are provided in Chinese, Japanese, Korean, Spanish, Portuguese, and more. - **Interactive Tools:** Answers can be checked, and patch notes are provided by version. - **Daily Farming Planner:** Saved on the player's device. ### Tech Stack: - Python ### Why It Matters: Irminsul provides accurate, source-cited answers to Genshin Impact questions, ensuring fair use without requiring an account and avoiding leaks or story spoilers.

Overview: NVIDIA Alpamayo 2 Edge is a 2B-parameter vision-language-action model designed for efficient in-vehicle deployment, generating driving trajectories and causal explanations using adaptive reasoning. Key features: - 2B-parameter model - Adaptive reasoning for fine-tuning and efficient deployment - Supports six-camera observations, ego history, and navigation instructions - Generates driving reasoning and 64 future poses at 0.1-second intervals Tech stack: PyTorch Why it matters: This model offers a compact yet powerful solution for in-vehicle applications, enabling real-time decision-making and trajectory generation with minimal computational overhead. Repository: NVlabs/alpamayo2-edge GitHub stars: 8 GitHub description: NVIDIA Alpamayo 2 Edge is a 2B-parameter reasoning vision-language-action model using adaptive reasoning to generate trajectories and causal explanations for fine-tuning and efficient deployment.

Overview: This project is an advanced RAG (Retrieval-Augmented Generation) pipeline designed to enhance text generation capabilities by integrating knowledge from multiple sources. Key features: - Integration of multiple sources for enhanced knowledge retrieval - Advanced text generation models - Customizable pipelines for various applications Tech stack: Jupyter Notebook Why it matters: This project offers a powerful tool for researchers and developers to improve the quality and relevance of their text generation tasks by leveraging diverse and accurate knowledge sources.

Overview: A lightweight implementation for running inference on local Large Language Models (LLMs). Key features: - Read and parse `.gguf` files - Simplified and optimized token generation Tech stack: Python Why it matters: This project provides a straightforward and efficient way to run inference on local hardware, making it useful for developers and researchers working with LLMs.

### Overview: A LangChain application designed to act as a Game Master for tabletop RPGs, integrating various fan-favorite settings and characters. ### Key Features: - **GUI Interface**: A user-friendly graphical interface for setting up a game, creating characters, and starting the adventure. - **CLI Commands**: Command-line interface for more advanced users, allowing for easier navigation and management of the game. - **Vector Stores**: Integration with Qdrant for efficient storage and retrieval of game-related data. - **Adaptive Memory**: The ability to remember game progress and context beyond the initial setting. ### Tech Stack: Python ### Why it Matters: This application streamlines the process of creating and managing a tabletop RPG, making it accessible and efficient for players and GMs alike.

Overview: TransAPI is an OpenAI compatible API purely based on Transformers, offering an alternative to unsupported models and models not available through GGUF. Key features: - OpenAI compatible API - Native integration with Transformers - Multimodal support (Text-to-Text and Image-Text-to-Text) - Reverse proxy support (e.g. Nginx) - Multiple sampling strategies including beam search - `torch.compile` for faster inference - `bitsandbytes` in-flight quantization - Various quantization formats supported by Transformers - Concurrent inference with asyncio - Streaming and non-streaming responses - Smart KV Cache offloading Tech stack: Python Why it matters: This project provides an alternative for unsupported models and models not available through GGUF, offering an OpenAI compatible API based entirely on Transformers, allowing you to run and test any model compatible with Transformers. Repository: Orion-zhen/transAPI Primary language: Python GitHub stars: 8 GitHub description: OpenAI compatible API purely based on Transformers

Overview: Tunebench is a CLI tool designed for developers to compare full fine-tuning, LoRA, and layer freezing techniques on small language models. Key features: - Full fine-tuning - LoRA (Layer-wise Adaptive Rate) - Layer freezing - Weight difference analysis - Catastrophic forgetting check - Gradient norm tracking - Layer-wise learning rate decay Tech stack: Python Why it matters: Tunebench simplifies the process of evaluating different fine-tuning strategies on small language models, making it easier for developers to understand and optimize their models. Repository: PanGan21/tunebench GitHub stars: 8 GitHub description: A CLI for comparing full fine-tuning, LoRA, and layer freezing on small LLMs.

### Overview: This repository provides a streamlined method to fine-tune the RF-DETR model on custom classes from the Open Images V7 dataset, making it accessible for developers looking to integrate object detection capabilities into their projects. ### Key Features: - Fine-tuning of RF-DETR on selected classes from Open Images V7 - Minimal setup required, leveraging Python and the UV package - Easy to use with a single command for fine-tuning - Supports fine-tuning on multiple classes simultaneously - Inference capabilities for new images ### Tech Stack: Python 3.10+ and UV package ### Why it Matters: This approach simplifies the process of integrating advanced object detection models into projects, making it accessible to developers without extensive machine learning expertise.

Overview: SpaceVector is a platform for semantic search on satellite images using state-of-the-art AI, aiming to support the use of satellite images. Key features: - Semantic search on satellite images - Integration with OpenAI Clip text and image encoders - Use of the EuroSat dataset for image embeddings - Docker and Docker Compose for easy deployment Tech stack: TypeScript Why it matters: SpaceVector leverages cutting-edge AI technology to enhance the accessibility and understanding of satellite imagery, making it a valuable tool for various applications in fields such as environmental monitoring, urban planning, and more. Repository: PoCInnovation/SpaceVector GitHub stars: 8 GitHub description: SpaceVector is a platform for semantic search on satellite images using state of the art AI. It aims to support the use of satellite images.

Overview: Periplus is an open-source in-memory vector database cache built on Meta's FAISS library, designed to store a dynamically updated subset of a large vector collection entirely in memory while serving queries without interacting with any other nodes at query time. Key features: - Dynamic vector database caching - In-memory storage - Supports FAISS library - Supports FAISS vector similarity search library Tech stack: C++ Why it matters: Periplus is an innovative solution for vector database caching, offering improved performance and lower latency compared to traditional methods.

Overview: Lorata is a data labeling tool designed specifically for GenAI models, offering a range of labeling tasks including text-to-image, image-to-image, and text-to-video. It also includes a built-in image/video editor that supports various editing functionalities. Key features: - Supports text-to-image, image-to-image, and text-to-video tasks. - Built-in image/video editor with features like drawing on images, cropping, trimming videos, generating captions, and extracting image frames from videos. - Export tasks as zip files with structured directories for easy management. Tech stack: TypeScript Why it matters: Lorata simplifies the process of labeling data for GenAI models, making it easier to train and improve these models.

### Overview: The AI-Powered Travel Booking Platform is a full-stack intelligent system designed for booking travel properties. It leverages RAG retrieval, multi-agent collaboration, and a smart pricing engine to provide a seamless and efficient booking experience. ### Key features: - 🏠 Property Browsing: Responsive grid layout for property browsing. - 🔍 Smart Search: Multi-criteria filtering for property search. - ❤️ Favorites: One-click save to favorites. - 📋 Listing Details: Complete info display and instant booking. ### Tech stack: - Frontend: Next.js 13 (App Router) · TypeScript · Tailwind CSS · Canvas 2D API - Backend: Next.js API Routes · Prisma ORM · MongoDB Atlas · NextAuth.js - AI/ML: Google Gemini Pro (NLP) · Gemini Embeddings (768-dim) · Supabase Vector (pgvector) · RAG Pipeline - Deployment: Vercel (Frontend + API) · MongoDB Atlas (Database) · Supabase (Vector DB) ### Why it matters: This platform is designed to revolutionize the travel booking industry by combining advanced AI and machine learning with intuitive user interfaces to provide a highly personalized and efficient booking experience.

### Overview: Model Context Protocol (MCP) Bootcamp is a comprehensive guide designed to provide a deep dive into the MCP architecture and its pivotal role within the Agentic AI ecosystem. The course focuses on practical, project-based implementations using LangChain, LangGraph, and CrewAI, ensuring learners can build real-world, production-ready AI workflows. ### Key Features: - **LangChain**: A powerful library for building and deploying AI models. - **LangGraph**: A graph-based AI platform for efficient data processing and analysis. - **CrewAI**: A framework for creating and deploying AI agents. - **Model Context Protocol (MCP)**: A foundational component for creating and managing AI models. - **LangChain MCP Adapters**: Customizable adapters for LangChain, enhancing its capabilities. - **Python**: The primary programming language used throughout the course. - **Google Gemini API**: Used for setting up the MCP server environment. - **Docker**: For creating and deploying MCP servers efficiently. - **Server-Sent Events (SSE)**: A method for real-time communication between the server and client. - **Multiple Server Support**: Allows for seamless integration of multiple MCP servers. - **Simplified Setup**: Easy-to-follow instructions for setting up and using MCP clients and servers. ### Tech Stack: The course utilizes the following technologies: - Python - LangChain - LangGraph - CrewAI - Model Context Protocol (MCP) - Google Gemini API - Docker - Server-Sent Events (SSE) - Multiple Server Support ### Why It Matters: This

### Overview: This project involves the manual fine-tuning of Sentence Transformer embeddings, focusing on improving the quality and performance of text embeddings for various applications. The fine-tuning process is designed to enhance the model's ability to handle specific domains or tasks, making it more suitable for real-world scenarios. ### Key Features: - Manual fine-tuning of Sentence Transformer embeddings. - Enhanced performance for specific domains or tasks. - Improved text embedding quality for various applications. ### Tech Stack: Jupyter Notebook ### Why It Matters: This project is crucial for developers looking to improve the accuracy and relevance of text embeddings in their applications. By fine-tuning Sentence Transformer embeddings manually, developers can tailor the model's performance to specific needs, ensuring better results in tasks such as text classification, summarization, and more.

Overview: LoLDU is a cutting-edge Parameter-Efficient Fine-Tuning (PEFT) technique designed to significantly reduce the number of trainable parameters while maintaining performance comparable to full fine-tuning. This project provides comprehensive documentation and implementation details for integrating LoLDU into various model architectures. Key features: - Significantly reduces the number of trainable parameters (up to 2600 times fewer than regular PEFT methods) - Maintains performance comparable to full fine-tuning - Leverages Lower-Diag-Upper Decomposition (LDU) for faster convergence and orthogonality - Focuses on optimizing the diagonal matrix for scaling transformations - Compatible with various model architectures (e.g., LLaMA2, Roberta, ViT, Stable Diffusion) Tech stack: Python, PyTorch, Jupyter Notebook Why it matters: LoLDU offers a significant improvement in parameter efficiency while preserving model performance, making it a valuable tool for fine-tuning models with limited computational resources. Repository: SKDDJ/LoLDU GitHub stars: 8 GitHub description: Official implementation for LoLDU: Low-Rank Adaptation via Lower-Diag-Upper Decomposition for Parameter-Efficient Fine-Tuning