Coding & Development

ai-edu is an open-source AI education platform developed by Microsoft Research Asia, specifically designed for Chinese students, teachers, and IT professionals. The platform aims to help learners master AI knowledge and develop practical skills through a structured curriculum. It comprises three main modules: basic tutorials, practical cases, and practical projects. The content covers fundamental AI concepts, neural network principles, classic machine learning algorithms, and modern software engineering. The platform is actively supported and updated by Microsoft Asia Research's R&D team and academic cooperation department, encouraging community contributions and feedback through GitHub issues and pull requests.

WanGP is an AI video application hosted on Hugging Face, designed to generate videos based on user-provided text prompts and images. Developed by DeepBeepMeep, this tool allows for creative experimentation in video production. To use WanGP, users are required to clone the Hugging Face Space, indicating a more hands-on approach to deployment and usage. While the Space is currently paused, its core functionality is centered around transforming textual descriptions and visual inputs into dynamic video content, making it suitable for those looking to explore AI-driven video creation.

Relari focuses on designing intelligence with intent, providing tools to transform ideas into thoughtful AI agents. Their flagship product, Nuvi, is an AI agent builder for Software 3.0, enabling users to turn natural language specifications into reliable and testable agents without needing to write code. Relari also supports the development of trustworthy AI through initiatives like Agent Contracts and Continuous Eval, ensuring AI systems behave as intended. This approach combines creativity with structure and intuition with rigor, resulting in AI that operates purposefully and reliably for various applications.

Awesome-Domain-LLM is a comprehensive open-source project designed to centralize and categorize large language models (LLMs), datasets, and evaluation benchmarks specifically tailored for vertical domains. This repository serves as a valuable resource for AI researchers and practitioners looking to apply LLMs to specialized industries such as healthcare, law, finance, education, and more. It includes a wide array of models, from general-purpose LLMs like LLaMA2 and ChatGLM3-6B, to domain-specific models like ChiMed-GPT for medicine, DISC-LawLLM for legal services, and Tongyi-Finance-14B for finance. The project also lists relevant datasets and robust evaluation benchmarks, facilitating the development, testing, and deployment of AI solutions across various sectors. Regular updates ensure the inclusion of new and enhanced models, datasets, and benchmarks, fostering continuous innovation in domain-specific AI applications.

DeepMedic is an open-source project offering a powerful system for Deep Learning-based segmentation of structures within biomedical 3D scans. It enables the straightforward creation and training of 3D Convolutional Neural Networks (CNNs) to identify and segment specific structures, provided ground truth labels are available for training. The system is designed to process NIFTI images, making it highly compatible with many biomedical tasks. Key features include support for TensorFlow 2.0.0, logging metrics to Tensorboard, on-the-fly input normalization, and augmentation via affine transforms. It is particularly useful for tasks like brain lesion segmentation in MRI scans and comes with examples and documentation to facilitate installation and use.

Cube Studio is an open-source, cloud-native, one-stop platform designed for machine learning, deep learning, and large AI models. It covers the full MLOps algorithm lifecycle, from online notebook development and drag-and-drop task flow pipeline orchestration to multi-machine, multi-card distributed training and hyperparameter search. The platform also provides inference service VGPU virtualization, edge computing, and automated annotation capabilities. It supports fine-tuning and training of large models like DeepSeek, VLLM, Ollama, and Mindie, along with private knowledge bases and an AI model market. Cube Studio is compatible with domestic CPUs/GPUs/NPUs (Ascend ecosystem), RDMA, and various distributed frameworks including PyTorch, TensorFlow, MXNet, DeepSpeed, Paddle, ColossalAI, Horovod, and Ray.

🤗 Spaces Semantic Search is an AI-powered tool designed to help users discover relevant Hugging Face Spaces. By simply typing a word or phrase into the search box, the application semantically searches through a large collection of Spaces, returning matching results. This functionality is particularly useful for individuals looking to explore new AI models, demos, and applications hosted on Hugging Face. The tool aims to streamline the discovery process, making it easier to navigate the vast ecosystem of AI resources available on the platform.

Math-Model-and-Machine-Learning is a comprehensive, open-source repository on GitHub, offering a wealth of notes and materials for individuals interested in mathematical modeling, machine learning, deep learning, and large models. Curated by an individual with a strong background in mathematics competitions, including a first prize in the Huawei Cup China Postgraduate Mathematical Modeling Contest, this project aims to support beginners. It includes resources such as competition problems, excellent papers, classic textbooks, and practical guides for machine learning, deep learning, and large models. The project is continuously updated and encourages community contributions to expand its content, making it a valuable learning hub for students and enthusiasts alike.

The MCP-Chinese-Getting-Started-Guide is an open-source resource designed to introduce developers to the Model Context Protocol (MCP). MCP is an innovative open-source protocol that standardizes how large language models (LLMs) interact with the external world, enabling seamless access and processing of information from diverse data sources and tools. This guide focuses on implementing MCP servers, particularly for integrating tools like web search, and demonstrates how to develop MCP clients to interact with these servers. It covers practical examples using Python 3.11, uv for project management, and includes debugging with the Inspector visualization tool. The guide also delves into advanced features like Sampling, which allows for human supervision during tool execution, enhancing control and safety.

Z Image Turbo LoRA DLC is a specialized image generation tool built on the Hugging Face Spaces platform, designed to leverage the Z-Image Turbo model. Users can input a text prompt and then select from a collection of impressive LoRAs (style models) provided in a gallery, or even upload their own custom LoRAs. The application then processes the input to generate a picture, applying the chosen style along with any specified size or seed settings. This tool is ideal for creative content generation, allowing for significant image customization and exploration of diverse artistic styles through its LoRA integration.

ZeroGPU-LLM-Inference is a powerful AI tool hosted on Hugging Face Spaces, offering a streaming LLM chat experience. Users can type questions or requests and receive immediate, written responses from a language model. A key feature is the optional web-search integration, which pulls short snippets from DuckDuckGo to enrich the model's responses. The application also provides controls for customizing the chat experience, allowing users to tailor interactions to their specific needs. This makes it a versatile tool for various conversational AI applications, from quick information retrieval to more in-depth discussions powered by real-time web data.

MLBox is a powerful Automated Machine Learning (AutoML) Python library designed to simplify and accelerate the development of machine learning models. It offers a comprehensive suite of features, including fast reading and distributed data preprocessing, cleaning, and formatting capabilities. The library also provides highly robust feature selection and leak detection, ensuring the quality and relevance of input data. For model optimization, MLBox includes accurate hyper-parameter optimization in high-dimensional spaces. It supports state-of-the-art predictive models for both classification and regression tasks, incorporating techniques like Deep Learning, Stacking, and LightGBM. Additionally, MLBox offers prediction with model interpretation, helping users understand the reasoning behind predictions.

text2vec is an open-source Python library designed for converting text into vector representations, a fundamental task in natural language processing. It provides implementations of various text embedding and text similarity calculation models, including Word2Vec, RankBM25, Sentence-BERT, CoSENT, and BGE. The tool enables users to transform words, sentences, and paragraphs into vector matrices, facilitating tasks like semantic matching and similarity computation. It supports both English and Chinese languages and offers pre-trained models for different use cases, including multilingual options. With features like multi-GPU/CPU inference and a command-line interface, text2vec is built for practical, out-of-the-box use in diverse NLP applications.

Blocks is an open-source framework built on top of Theano, designed to simplify the construction and training of neural networks. It offers several key features including the ability to create 'bricks' for parametrized Theano operations, pattern matching for selecting variables and bricks within complex models, and algorithms for model optimization. The framework also supports saving and resuming training sessions, monitoring and analyzing training progress across different datasets, and applying graph transformations like dropout. Blocks is complemented by Fuel, a data processing engine, and has additional components available through Blocks-extras, making it a comprehensive solution for deep learning development.

awesome-feature-engineering is a comprehensive, curated list of resources dedicated to various feature engineering techniques essential for machine learning. This open-source repository covers a wide array of data types, including numeric, textual, image, categorical, time series, and geospatial data. It provides links to relevant libraries, articles, and tutorials for methods such as scaling, ranking, quantization, Box-Cox transformation, feature interactions, clustering, t-SNE, PCA, Bag of Words, TFIDF, word embeddings, one-hot encoding, count encoding, label encoding, mean encoding, hashing, rolling window features, and lag features. Maintained by Andrei Khobnia, this resource is invaluable for data scientists and machine learning engineers looking to enhance their feature engineering skills and find practical implementations.

Awesome-DynamicGraphLearning is a comprehensive, open-source GitHub repository dedicated to collecting and organizing significant research papers and their associated code in the field of machine learning, specifically deep learning, applied to dynamic (temporal) graphs, networks, and knowledge graphs. The repository covers a wide range of topics, including surveys, theoretical advancements, and applications such as recommender systems. It features papers from top conferences and journals like ICML, SIGKDD, ICLR, NeurIPS, WWW, and VLDB, spanning from 2012 to 2025. This curated list serves as an invaluable resource for researchers, academics, and students looking to stay updated on the latest developments and find relevant implementations in dynamic graph learning.

The AWS Machine Learning University: Accelerated Tabular Data Class offers a comprehensive open-source curriculum designed to introduce individuals to machine learning techniques specifically for tabular data. This repository provides a rich set of educational materials, including detailed slides, interactive notebooks, and real-world datasets. The course covers essential topics such as exploratory data analysis, K-Nearest Neighbors, feature engineering, tree-based models, boosting, and neural networks. It aims to make machine learning accessible to a broad audience, enabling learners to apply these techniques to practical problems. The curriculum culminates in a final project that allows students to practice working with a real-world tabular dataset.

AutoDidact is an open-source project designed to autonomously train research-agent LLMs on custom data. It leverages reinforcement learning and self-verification to enable small LLMs, such as Llama-8B, to enhance their research and reasoning capabilities. The tool allows LLMs to generate, research, and answer self-created question-answer pairs, learning agentic search through Group Relative Policy Optimization (GRPO). It features an entirely autonomous pipeline, covering question generation, answer research, verification, embedding creation, and reinforcement learning, all running locally on open-source models. Demonstrated results show significant accuracy improvements in research and question answering after minimal training, making it a powerful tool for developers and researchers looking to build self-improving AI agents.

AQLM is an official PyTorch repository for extreme compression of large language models (LLMs) through additive quantization. This open-source tool significantly reduces the size of LLMs while maintaining or improving accuracy, making them more efficient for deployment and inference. It supports various models from the LLaMA, Mistral, and Mixtral families, offering pre-quantized models and a finetuning algorithm called PV-tuning. AQLM provides inference kernels optimized for both GPU and CPU, with considerable speedups for specific quantization schemes. The repository includes detailed instructions for installation, model quantization, and evaluation, making it a valuable resource for researchers and developers focused on LLM optimization.

Dreambooth Submission is an AI tool hosted on Hugging Face, designed for image generation. It leverages the Keras-Dreambooth framework, indicating its foundation in deep learning for creating custom image models. The tool is tagged for 'wildcard' use, suggesting flexibility in its application, and is specified for the US region. While the live website content indicates a runtime error, the underlying purpose is to enable users to submit data and generate images through AI, likely for personalized or specific visual content creation. Its integration with Hugging Face Spaces makes it accessible within that ecosystem.

Datacadabra BV specializes in providing AI as a Service, developing artificial intelligence solutions to make tasks smarter and more efficient for businesses. The company focuses on leveraging data and AI to optimize and simplify work processes, enabling efficient and effective workforce deployment. Their expertise spans various sectors, including improving biodiversity with technologies like MowHawk in outdoor spaces, applying AI in the medical sector, and optimizing waste streams. Datacadabra offers workshops and case studies to demonstrate their methodology and the practical applications of their AI technology.

golearn is a comprehensive machine learning library designed for the Go programming language, emphasizing both simplicity and customizability. It offers a 'batteries included' approach, providing a wide range of functionalities for machine learning tasks. Users can load data as Instances, perform matrix-like operations, and pass them to various estimators. The library implements the scikit-learn interface of Fit/Predict, allowing for easy swapping of estimators during trial and error. Additionally, golearn includes helper functions for data management, such as cross-validation and train-test splitting. It supports various algorithms including KNN, linear models, neural networks, and decision trees, making it suitable for diverse machine learning applications.

prismatic-vlms offers a flexible and efficient codebase for training visually-conditioned language models (VLMs). It natively supports diverse visual backbones like CLIP, SigLIP, and DINOv2, with an easy mechanism for adding new ones via TIMM. The tool also integrates with arbitrary instances of AutoModelForCausalLM from Transformers, including both base and instruct-tuned language models. Designed for easy scaling, prismatic-vlms leverages PyTorch FSDP and Flash-Attention to efficiently train models ranging from 1B to 34B parameters on configurable dataset mixtures. It also includes an evaluation codebase for rigorously testing VLMs across 12 vision-and-language benchmarks and provides full instructions and configurations for reproducing results.

pytorch-attention offers a robust PyTorch implementation of various cutting-edge deep learning models, including a wide array of attention mechanisms, vision transformers, MLP-like models, and convolutional neural networks. This open-source codebase is designed for researchers and engineers to easily experiment with and integrate advanced architectures into their projects. It features implementations of models like Squeeze-and-Excitation Attention, ViT, ResNet, and MLP-Mixer, complete with code examples for quick setup and testing. The repository is modular and extensible, making it a valuable resource for anyone working on computer vision and deep learning tasks, providing a foundation for both academic research and practical application development.