Projects

### Overview: GUNDAM is a data management system that leverages language models to prioritize and manage textual data efficiently. Built on PyTorch, GUNDAM includes a comprehensive data manager with a GPT-2 generator and a demonstration retriever, making it highly flexible and efficient. ### Key features: - **Comprehensive Data Manager**: Includes a one-to-one miner, GPT-2 generator, and demonstration retriever. - **Flexible Support**: Supports GPT-2 models of various sizes. - **Efficient Data Quality Check**: Provides an efficient one-to-one miner for data quality assessment. ### Tech stack: Python ### Why it matters: GUNDAM is a groundbreaking tool that enhances the efficiency and effectiveness of data management by leveraging language models, making it indispensable for applications requiring data prioritization and demonstration management.

### Overview: SubgraphRAG is a simple yet effective retrieval-and-reasoning pipeline designed for knowledge-graph-based retrieval-augmented generation, showcasing the roles of graphs and large language models in this context. ### Key Features: - Simple yet effective retrieval and reasoning pipeline. - Supports knowledge-graph-based retrieval-augmented generation. - Integrates the strengths of graphs and large language models for improved performance. ### Tech Stack: Python ### Why It Matters: This project demonstrates the practical application of combining knowledge graphs with large language models to enhance retrieval-augmented generation, offering a simple yet effective solution for knowledge-based generation tasks.

Overview: The AI Horde Worker reGen is a software client designed for volunteers to generate images for the crowdsourced AI-Horde platform. By running a worker on your local machine, you can earn kudos, which are used to prioritize your own requests. Key features: - Generate images from text prompts - Earn kudos for completing jobs - Support for Windows, Linux, AMD GPUs, and DirectML Tech stack: Python Why it matters: The AI Horde Worker reGen allows individuals to contribute their computing power to the AI-Horde project, generating images and helping to advance the technology. By running the worker, you can earn kudos that can be used to expedite your own requests on the platform. Repository: Haidra-Org/horde-worker-reGen Primary language: Python GitHub stars: 187

Overview: Blenderwright is an intuitive MCP server for Blender that guides LLMs like Claude to produce professional 3D results through expert prompts and proven workflows, all through natural language. Key features: - 276 tools across 32 modules covering modeling, mesh editing, materials, shader nodes, lighting, camera, animation, rendering, sculpting, UV mapping, physics, geometry nodes, rigging, curves, sweeps, 3D-print checking, annotations, collections, file I/O, Bool Tool, viewport control, mesh quality analysis, and extension suggestions. - 13 expert prompts for topology best practices, real-world scale references, lighting principles, studio setup, character basemesh workflow, PBR material guide, auto-critique feedback loop, and more. - Visual feedback loop with fast viewport screenshots via OpenGL render (~ms, not seconds) and auto-critique prompts that guide the LLM to check its own work. - Mesh quality analysis with structured reports covering non-manifold edges, loose vertices, zero-area faces, duplicate vertices, and wire edges. - Extension suggestions proactively recommend Bool Tool, LoopTools, and Node Wrangler when a task would benefit from them. - Raw Python is opt-in: `execute_blender_code` is off until "Allow raw Python" is ticked in the N-panel, per session, never saved into the .blend. When on, every snippet is echoed to the console and listed in the panel. A restricted namespace blocks obvious mistakes (`os`, `subprocess`, `exec`, `open`, ...);

### Overview: PII Masker is an open-source tool designed to protect sensitive data by automatically detecting and masking Personally Identifiable Information (PII) using advanced AI models. Powered by DeBERTa-v3, it offers high-precision detection, scalable performance, and a simple Python API, ensuring privacy compliance in various industries. ### Key Features: - **Comprehensive Protection**: Detects and masks multiple PII types including names, addresses, phone numbers, and more. - **High Performance**: Utilizes DeBERTa-v3 with 1024 token support for processing longer documents. - **Precision Focused**: Fine-tuned AI model for PII detection. - **Structured Output**: Provides both masked text and a structured PII dictionary. - **Easy Integration**: Simple Python API for seamless integration into workflows. ### Tech Stack: - **Python**: 3.8+ - **Hugging Face Models**: Models fine-tuned for PII detection. - **Milvus**: Vector database for structured data storage. ### Why It Matters: - **High Precision**: Ensures accurate detection and masking of PII. - **Compliance Friendly**: Designed to meet privacy regulations. - **Flexible Integration**: Easy integration with existing systems through a Python API. ### Tech Stack: - **Python**: 3.8+ - **Hugging Face Models**: Models fine-tuned for PII detection. - **Milvus**: Vector database for structured data storage. ### Why Choose PII Masker? - **Comprehensive Protection**: Identifies and masks various PII types.

### Overview: Dynamic Parametric Retrieval Augmented Generation for Test-time Knowledge Enhancement (DyPRAG) is a novel framework that leverages a lightweight parameter translator model to efficiently map documents into parameterized knowledge. This approach reduces inference, training, and storage costs while enhancing language models' knowledge at test-time, specifically addressing the issue of RAG hallucination. ### Key Features: - **Parameterized Knowledge Mapping**: DyPRAG models the relationship between documents and parameters, enabling efficient knowledge representation. - **Test-time Knowledge Enhancement**: It dynamically integrates parameterized knowledge to resolve conflicts between contextual and parametric knowledge, providing a practical solution to mitigate RAG hallucination. - **Parameter Translator Model**: A lightweight model is used to translate documents into parameters, reducing the computational overhead. - **Extensive Experiments**: DyPRAG has been extensively tested on multiple datasets, demonstrating its effectiveness and generalization in test-time knowledge enhancement. ### Tech Stack: - Python ### Why It Matters: DyPRAG offers a practical and effective solution to mitigate RAG hallucination in real-world applications, making it a valuable addition to the toolkit of language model developers and researchers.

### Overview: Hyper-Diffusion-Planner is an official implementation of "Unleashing the Potential of Diffusion Models for End-to-End Autonomous Driving." This project showcases how diffusion models can effectively serve as planners for complex, real-world autonomous driving tasks. ### Key Features: - **Real Vehicle Demo**: Demonstrates the performance of the model on real-world urban scenarios. - **Nuplan Implementation**: Provides a benchmark implementation based on the Nuplan platform. - **NAVSIM Implementation**: Offers a benchmark implementation based on the NAVSIM devkit, featuring supervised diffusion VLA and reward-based fine-tuning. ### Tech Stack: - Python ### Why It Matters: This project highlights the potential of diffusion models in autonomous driving, showcasing their effectiveness and scalability for end-to-end autonomous driving tasks.

Overview: MUXI is an open-source AI application server designed for deploying AI agents in production environments. Key features: - Native AI agents as primitives - Self-hosted server stack with orchestration - Auditable memory management - Group-based RBAC - Twelve SDKs already integrated - Comprehensive formation capabilities Tech stack: MUXI Why it matters: MUXI simplifies the process of deploying AI agents by providing a complete, self-hosted server stack, eliminating the need to reinvent infrastructure. It offers production capabilities through native AI agents, ensuring reliability and scalability.

Overview: The AdEMAMix Optimizer is a state-of-the-art optimization algorithm designed to be both faster and more effective than existing methods. Key features: - **Improved Convergence Speed**: Significantly reduces the time required to converge to optimal solutions. - **Enhanced Performance**: Offers better performance across various optimization problems. - **Older but Better**: Built on the foundation of older algorithms but with modern improvements. Tech stack: Python Why it matters: The AdEMAMix Optimizer is a significant advancement in optimization algorithms, offering faster and more efficient solutions to complex problems, making it invaluable for researchers and developers alike. Repository: nanowell/AdEMAMix-Optimizer-Pytorch Primary language: Python GitHub stars: 187

### Overview: Quixotic AI is a complete, open-stack solution for local AI inference on the JVM, designed from first principles. It provides a unified API for various AI tasks and runs end-to-end on the JVM, eliminating the need for sidecar servers, ONNX, or external services. ### Key features: - **AI inference engine**: A complete inference engine for chat, vision, audio, embeddings, reranking, and text-to-speech. - **Token-perfect tokenizers**: Fast tokenizers for large language models, designed for perfect tokenization. - **Native matrix multiplication**: Optimized matrix multiplication routines for CPUs and GPUs, achieving performance comparable to `llama.cpp`. - **GraalVM's Native Image**: First-class support for GraalVM Native Image, resulting in self-contained binaries with a small footprint and fast startup times. ### Tech stack: Quixotic AI is built on Java, leveraging the GraalVM ecosystem for efficient matrix multiplication and native image compilation. ### Why it matters: Quixotic AI offers a pure Java-based inference solution that runs on the JVM, eliminating the need for Python, ONNX, or external services. This makes it ideal for applications that require high performance and security, such as mission-critical systems and enterprise environments.

Overview: MCP Image Generator is a server-side tool designed for AI image generation and editing, optimized for prompt optimization and quality presets. It supports Nano Banana (Gemini), OpenAI GPT Image, and BytePlus Seedream, making it versatile for various image generation needs. Key features: - Automatic prompt optimization - Supports multiple image providers: Nano Banana (Gemini), OpenAI GPT Image, and BytePlus Seedream - Enhances image quality and composition based on the provided details Tech stack: Node.js 22 or later, TypeScript Why it matters: MCP Image Generator streamlines the image generation process by enhancing prompts, ensuring high-quality output, and supports multiple providers, making it a valuable tool for developers and creators looking to automate their image generation workflows. Repository: shinpr/mcp-image Primary language: TypeScript GitHub stars: 185 GitHub description: MCP server for AI image generation and editing with automatic prompt optimization and quality presets. Supports Nano Banana (Gemini), OpenAI GPT Image, and BytePlus Seedream.

Overview: This project embeds multiple modalities into large language models, including images, audio, and documents, to enhance their capabilities. Key features: - Embeds images, audio, and documents into large language models. - Supports various modalities and models. - Uses flash-attn for efficient computation. Tech stack: Python Why it matters: This library expands the functionality of large language models by integrating diverse data types, potentially opening new possibilities in AI applications. Repository: sshh12/multi_token Primary language: Python GitHub stars: 188 GitHub description: Embed arbitrary modalities (images, audio, documents, etc) into large language models.

### Overview: This repository provides the official implementation of the paper "Self-correcting LLM-controlled Diffusion Models (SLD)" at CVPR 2024. The project is designed to enhance generative models by integrating LLM detectors for precise text-to-image alignment, making it suitable for both image generation and fine-grained editing. ### Key Features: - **Self-correction**: Enhances generative models with LLM-integrated detectors for precise text-to-image alignment. - **Unified Generation and Editing**: Excels at both image generation and fine-grained editing. - **Universal Compatibility**: Works with any image generator, such as DALL-E 3, requiring no extra training or data. ### Tech Stack: The project uses Python with the PyTorch framework. ### Why It Matters: SLD offers a robust solution for improving generative models by integrating LLM detectors, making it versatile and compatible with various image generators, enhancing both image generation and fine-grained editing capabilities.

Overview: TiMEM-AI/TiMEM is a Predictive Cognitive Regulation Engine powered by Temporal-Hierarchical Memory (TMT). It enables agents to handle long-horizon tasks with stable execution, continuous self-evolution, and proactive user understanding. Key features: - **5-Level Temporal Memory Tree (TMT)**: Explicit temporal ordering from fine-grained evidence to stable persona. - **Predictive Cognitive Regulation**: Proactively anticipates agent needs and user intent via temporal-hierarchical memory. - **Continuous Self-Evolution**: Agents learn and adapt from every interaction without fine-tuning. - **Long-Horizon Task Stability**: Sustained performance across extended sessions and complex workflows. - **Complexity-Aware Recall**: Adaptive retrieval that balances precision and efficiency. Tech stack: Python. Why it matters: TiMEM is a game-changer for agents tasked with handling long-horizon tasks, ensuring stable execution, continuous learning, and proactive user understanding. Its predictive cognitive regulation and temporal memory tree (TMT) capabilities make it a state-of-the-art solution for enhancing the performance of conversational agents and autonomous agents alike. Repository: TiMEM-AI/TiMEM Primary language: Python GitHub stars: 186 GitHub description: Temporal-Hierarchical Memory Consolidation for Long-Horizon Conversational Agents README excerpt: [...] GitHub description: Temporal-Hierarchical Memory Consolidation for Long-Horizon Conversational Agents Overview: TiMEM-AI/TiMEM is a Predictive Cognitive Regulation Engine powered by Temporal-Hierarchical Memory (T

### Overview: QuicKB optimizes document retrieval by fine-tuning knowledgebases through an end-to-end machine learning pipeline that handles document chunking, synthetic data generation, and embedding model training. ### Key Features: - **Document Chunking**: Implements state-of-the-art chunking strategies based on ChromaDB's research, including semantic approaches and token-based methods. - **Training Data Generation**: Automatically creates domain-specific training datasets using synthetic question-answer pairs, semantic similarity, and parallel processing. - **Embedding Optimization**: Fine-tunes embedding models for specific domains, cross-platform support, dimension reduction, and comprehensive evaluation metrics. ### Tech Stack: Python ### Why it Matters: QuicKB enhances document retrieval efficiency by leveraging advanced machine learning techniques, ensuring optimal performance and scalability.

### Overview: KB-NER is a knowledge-based system that achieved top results in the SemEval 2022 MultiCoNER shared task, ranking 10 out of 13 tracks. The system leverages a multilingual knowledge base built on Wikipedia to enhance the accuracy of complex named entity recognition. ### Key features: - **Knowledge Base Building**: KB-NER constructs a multilingual knowledge base using Wikipedia to provide contextual information for named entity recognition. - **Retrieval-based Data Augmentation**: The system uses retrieval-based data augmentation techniques to improve model performance. - **Context Processing**: KB-NER processes the context around named entities to enhance the accuracy of the recognition. - **Multi-stage Fine-tuning**: The model undergoes multiple stages of fine-tuning to improve its performance. - **Majority Voting Ensemble**: KB-NER employs a majority voting ensemble approach to improve accuracy. ### Tech stack: Python ### Why it matters: KB-NER's ability to leverage a multilingual knowledge base and advanced data augmentation techniques has significantly improved the accuracy of named entity recognition across various languages, making it a valuable tool for researchers and developers working on natural language processing tasks. ### Repository: Alibaba-NLP/KB-NER ### Primary language: Python ### GitHub stars: 185 ### GitHub description: Winner system (DAMO-NLP) of SemEval 2022 MultiCoNER shared task over 10 out of 13 tracks. [[Rankings]](https://multiconer.github.io

Overview: This Python wrapper provides access to Black Forest Labs' FLUX 3 Dev API, offering a faster, lower-cost variant of the FLUX 3 model. It includes Text-to-Image, Image-to-Image, Text-to-Video, and Image-to-Video functionalities, with the flagship Text-to-Image endpoint being live now. Key features: - Text-to-Image - Image-to-Image - Text-to-Video - Image-to-Video Tech stack: Python Why it matters: This SDK simplifies the integration of FLUX 3 Dev into projects, enabling rapid iteration and prototyping with reduced latency and cost compared to the full FLUX 3 model.

### Overview: MedXpertQA is a highly challenging and comprehensive benchmark designed to evaluate expert-level medical reasoning and understanding. It includes 4,460 questions spanning 17 specialties and 11 body systems, with two subsets: MedXpertQA Text for text medical evaluation and MedXpertQA MM for multimodal medical evaluation. This benchmark aims to provide a realistic and challenging environment for evaluating the performance of AI models in medical reasoning tasks. ### Key Features: - **Next-Generation Multimodal Medical Evaluation:** MedXpertQA MM introduces expert-level medical exam questions with diverse images and rich clinical information, including patient records and examination results, setting it apart from traditional medical multimodal benchmarks with simple QA pairs generated from image captions. - **Highly Challenging:** MedXpertQA introduces high-difficulty medical exam questions and applies rigorous filtering and augmentation, effectively addressing the insufficient difficulty of existing benchmarks like MedQA. The Text and MM subsets are currently the most challenging benchmarks in their respective fields. - **Clinical Relevance:** MedXpertQA incorporates specialty board questions to improve clinical relevance and comprehensiveness by collecting relevant medical data and integrating them into the evaluation process. ### Tech Stack: Python ### Why It Matters: MedXpertQA provides a realistic and challenging environment for evaluating AI models in medical reasoning tasks, making it an essential benchmark for researchers and developers in the field.

Overview: TriviumDB is a lightweight embedded database engine that seamlessly fuses vector search, graph, and document-oriented metadata within a single storage kernel. It is designed to be battle-tested in mission-critical, air-gapped environments, offering a highly efficient and secure solution for AI applications. Key features: - Vector search (Vector) - Graph (Graph) - Document-oriented metadata (Document) Tech stack: Rust Why it matters: TriviumDB simplifies the development of AI applications by providing a unified database solution that integrates vector search, graph processing, and document metadata. This eliminates the need for separate databases and simplifies the development process, making it an essential tool for AI developers looking to build robust and efficient applications.

Overview: This roadmap provides a beginner's path through the essential skills required for AI engineering in 2026, focusing on the shift from "train models" to "build reliable agent systems" and the three skills (RAG, Agents, and Evaluation) that appear in every job description. Key features: - Foundations - LLM Engineering - Agents - Evaluation - Production (LLMOps) Tech stack: None specified Why it matters: This roadmap is designed to help aspiring AI engineers navigate the rapidly evolving landscape of AI engineering in 2026, emphasizing the skills needed to build reliable agent systems and the importance of evaluation engineering. Repository: h9-tec/ai-roadmap-2026 GitHub stars: 178 GitHub description: AI Roadmap 2026: a beginner's path through Foundations, LLM Engineering, Agents, Evaluation, and Production, with verified free resources and a mini-project per layer.