Coding & Development

awesome-gpt is an extensive, open-source collection of resources dedicated to ChatGPT, offering a diverse array of tools, documents, applications, and practical use cases. This GitHub repository serves as a central hub for anyone looking to explore or integrate ChatGPT into their projects. It features resources categorized by programming languages like Python, Go, Kotlin, and JavaScript, alongside sections for API tools, client-side implementations, browser extensions, desktop applications, and editor integrations. The collection also highlights various chat bots, web applications, and CLI tools, making it a valuable reference for developers, researchers, and enthusiasts seeking to leverage the capabilities of ChatGPT across different platforms and applications. Contributions and suggestions are actively welcomed, fostering a continuously growing and up-to-date resource.

Awesome-Controllable-Diffusion is a comprehensive, open-source repository dedicated to papers and resources focused on controllable generation using diffusion models. This curated list is invaluable for researchers and developers working in the field of AI-generated content (AIGC). It covers significant advancements and techniques, including ControlNet, DreamBooth, and IP-Adapter, offering a centralized hub for staying updated on the latest research. The repository is meticulously organized by year, making it easy to navigate and find relevant academic papers and associated codebases for various controllable diffusion model applications.

smalldiffusion is a lightweight, open-source Python library designed for training and sampling from diffusion and flow models. It prioritizes ease of experimentation, allowing developers and researchers to quickly train new models or develop novel samplers. The library supports a variety of models including MLP, U-Net, and DiT, and multiple parameterizations such as score-, flow-, or data-prediction. It offers dataset support for 2D toy datasets, pixel, and latent-space image datasets, with example training code for FashionMNIST, CIFAR10, and Imagenet. smalldiffusion also provides concise implementations of diffusion transformers and supports conditional training with classifier-free guidance, making it a versatile tool for those working with diffusion models.

MetaY is a groundbreaking Chrome extension that empowers users to contribute their unused GPU resources for AI deep learning inference tasks. By leveraging the excess capacity of your GPU, MetaY enables you to earn rewards while supporting cutting-edge AI research and development. The app is simple to set up and run, ensuring you make the most out of your GPU without compromising your own usage. It features optimized GPU utilization, a secure and transparent operation, and a user-friendly interface to monitor contributions and track earned points. MetaY is ideal for tech enthusiasts, AI researchers, gamers, and GPU owners looking to monetize their idle computing resources.

Awesome-Multimodal-Large-Language-Models is a GitHub repository dedicated to cataloging the most recent developments in multimodal large language models (LLMs). It serves as a valuable resource for anyone interested in the intersection of large language models with various modalities like vision, audio, and more. The repository organizes links to surveys, research papers, and ongoing projects, offering a structured overview of the field. It specifically highlights topics such as instruction-following, in-context learning, chain-of-thought reasoning, and instruction-tuning within multimodal contexts. This makes it an essential reference for researchers, academics, and practitioners looking to stay updated on unified multimodal understanding and generation.

awesome-rnn offers a comprehensive, curated list of resources focused on recurrent neural networks (RNNs), a key area within deep learning. This GitHub repository serves as a central hub for researchers and students, providing links to various implementations in frameworks like TensorFlow, Theano, Torch, and PyTorch. It also categorizes resources by theory, lectures, books, and architecture variants such as LSTM and GRU. Furthermore, it details applications across natural language processing, computer vision, and robotics, making it an invaluable reference for understanding and implementing RNNs.

Awesome-Rust-MachineLearning is a comprehensive repository listing machine learning libraries written in Rust. It serves as a compilation of GitHub repositories, blogs, books, movies, discussions, and papers, specifically targeting individuals considering a migration from Python to Rust for machine learning tasks. The repository categorizes libraries by basic functionality and algorithm types, including support tools like Jupyter Notebook integration, graph plotting, vector and dataframe manipulation, image processing, and natural language processing. It also covers advanced topics such as GPU acceleration, comprehensive ML frameworks like SmartCore and Linfa, gradient boosting, deep neural networks, and reinforcement learning. The resource also includes libraries that may no longer be actively maintained, offering a broad overview of the Rust ML ecosystem.

devclad is an AI-powered platform specifically crafted for the developer community, aiming to enhance networking and collaboration. It offers a dedicated environment where developers can build software, share their code, and engage with peers on various projects. The platform's core mission is to foster a vibrant ecosystem of collaboration and innovation, enabling developers to connect, learn from each other, and collectively advance their work. While specific features are not detailed, the emphasis is on creating a supportive and interactive space for technical professionals.

diffusion_models offers a series of tutorial notebooks designed to educate users on denoising diffusion probabilistic models (DDPMs) using PyTorch. These tutorials delve into the theoretical underpinnings of DDPMs, drawing inspiration from thermodynamics and denoising score matching. The resource also covers related topics such as Langevin dynamics, autoregressive decoding, and the more recent denoising diffusion implicit models (DDIMs) for accelerated sampling. Additionally, it explores practical applications like the WaveGrad model for audio data. The notebooks are available in both PyTorch and JAX, providing a comprehensive learning experience for those interested in advanced generative AI models.

Awesome-LLM is a comprehensive, curated list of resources focused on Large Language Models (LLMs), particularly those related to ChatGPT. It serves as an invaluable resource for researchers, developers, and enthusiasts in the AI community. The repository categorizes information into key areas such as milestone papers, other relevant papers, LLM leaderboards, data sets, evaluation metrics, training frameworks, inference techniques, applications, tutorials, courses, and books. It also highlights trending LLM projects and offers insights into various subfields of LLM research, making it a central hub for staying updated on the latest advancements and foundational knowledge in the LLM space.

Browserable is an open-source and self-hostable browser automation library specifically designed for AI agents. It empowers developers to create intelligent agents capable of navigating websites, interacting with web elements like forms and buttons, and extracting valuable information. The library boasts a strong performance, achieving 90.4% on the Web Voyager benchmarks, indicating its effectiveness in complex web automation tasks. It offers flexible configuration options for LLM providers, storage solutions, database systems, remote browsers, and custom functions. Browserable provides a JavaScript SDK for easy integration and offers various services including a UI server, documentation, task management API, and database management tools, making it a comprehensive solution for AI-driven web interaction.

ChatterBot is an open-source, machine-learning based conversational dialog engine implemented in Python, designed for creating chat bots. It operates by generating responses from collections of known conversations, making it highly adaptable. The tool's language-independent architecture allows it to be trained in any language. Initially, ChatterBot has no knowledge of communication, but it learns by saving user input and corresponding responses. As it receives more data, its ability to reply accurately increases. It selects the closest matching response by finding the most similar known statement to the input and then returns the most likely response based on frequency of use. Documentation and training data for over a dozen languages are available, with contributions welcomed.

DeepSNAP is a Python library designed to facilitate efficient deep learning on graphs. It offers robust support for flexible graph manipulation, integrating with powerful graph libraries like NetworkX and deep learning frameworks such as PyTorch Geometric. The library provides a standard pipeline for tasks like dataset splitting, negative sampling, and defining node/edge/graph-level objectives, ensuring transparency for users. DeepSNAP also efficiently supports flexible and general heterogeneous Graph Neural Networks (GNNs), accommodating both node and edge heterogeneity. Its intuitive API allows users to control message parameterization and passing, making it easy to use for those familiar with PyTorch Geometric.

GPT-3, developed by OpenAI, is a powerful language model that enables developers to integrate advanced AI capabilities into their applications via an API. It excels at generating human-like text completions from natural language prompts, making it suitable for tasks like search, conversation, and text completion. With over 300 applications leveraging its capabilities and processing an average of 4.5 billion words per day, GPT-3 is a cornerstone for innovative AI development. The platform offers features like Answers and Classifications endpoints for building intelligent bots and autoML solutions, alongside robust safety measures and a prompt library to aid developers in creating diverse applications across various industries.

RedPajama-Data is an open-source repository containing code designed to prepare extensive datasets for training large language models. This tool facilitates the creation and management of high-quality training data through a multi-step pipeline. Key functionalities include preparing artifacts like quality classifiers and generative models, computing various quality signals such as perplexity scores and importance weights, and performing both exact and fuzzy deduplication to refine the dataset. It supports multiple languages including English, German, French, Italian, and Spanish, and offers a robust framework for researchers and developers working with large-scale language model training.

SLAM-LLM is a comprehensive deep learning toolkit designed for researchers and developers to train custom multimodal large language models (MLLMs). It specializes in processing speech, language, audio, and music, offering detailed recipes for training and high-performance checkpoints for inference. The framework supports multi-task training, dynamic prompt selection, and iterative datasets for large-scale industrial applications, including datasets on the order of 100,000 hours. Key features include DeepSpeed training for reduced memory usage, multi-machine multi-GPU inference, and dynamic frame batching to significantly reduce training and evaluation times. It also provides flexible configuration options based on Hydra and dataclass, allowing for a combination of code, command-line, and file-based configurations.

sod is an embedded, modern cross-platform computer vision and machine learning software library. It offers a comprehensive set of APIs for deep-learning, advanced media analysis, and real-time, multi-class object detection, even on systems with limited computational resources and IoT devices. Built for computational efficiency, sod includes both classic and state-of-the-art deep-neural networks with pre-trained models, including its exclusive RealNets architecture. It is dependency-free, written in C, and compiles into a single C file for easy deployment across various platforms. Use cases range from real-time object detection and facial recognition to license plate extraction and intrusion detection.

Qualcomm AI Hub is designed to streamline the deployment of artificial intelligence models to edge devices. It focuses on optimizing and validating AI models to ensure high-performance and low-power computing, which is crucial for edge AI applications. The platform enables developers to efficiently deploy their AI models on various Qualcomm platforms, facilitating the integration of advanced AI capabilities into a wide range of devices. This tool is particularly valuable for those working with embedded systems and IoT devices where computational efficiency and power consumption are key considerations.

Cortica is a pioneer in Autonomous AI, having invested over $250M and secured 300+ patents over 15 years to develop its groundbreaking technology. Its revolutionary AI mirrors how the human cortex processes information, utilizing signatures for generic representations, adaptive architecture for scenario-focused adaptivity, and self-learning neural networks independent of manually labeled data. This technology enables efficient data processing, superior performance, and scalability on low-compute platforms. Cortica partners with global market leaders to build AI companies like Qualisense (quality inspection), Autobrains (autonomous vehicles), Corsight (facial recognition), SeeTrue (threat detection), CORDiguide (cardiovascular procedures), and Corsound (voice biometrics), providing them with a technological and business advantage in large market opportunities.

DreamPose is an official implementation of "DreamPose: Fashion Image-to-Video Synthesis via Stable Diffusion." This open-source project allows users to generate dynamic videos from static fashion images by leveraging pretrained models. The tool supports finetuning on custom datasets, enabling the creation of subject-specific models. Users can prepare data by creating train and test subdirectories, running DensePose for pose estimation, and reformatting the output. DreamPose is built upon a pretrained Stable Diffusion checkpoint and can be finetuned on image datasets, with instructions provided for training on NVIDIA A100 GPUs. It also offers options to finetune the UNet and VAE decoder components for customized video generation.

Anomaly Detection is a Hugging Face Space by sklearn-docs that provides an interactive platform to understand and visualize anomaly detection algorithms. Users can select different datasets and adjust parameters like the number of samples and the fraction of outliers to observe how various algorithms identify anomalies. This tool is particularly useful for those interested in machine learning and data analysis, offering a hands-on approach to explore the effectiveness of different anomaly detection techniques. Although currently paused, its purpose is to demonstrate the capabilities of scikit-learn's anomaly detection methods in a user-friendly environment.

TrojAI is an AI security platform designed to protect AI models and applications throughout their lifecycle. It offers two core products: TrojAI Detect, which secures AI models at build time by identifying weaknesses and vulnerabilities before deployment, and TrojAI Defend, which safeguards AI applications at runtime against real-time threats. The platform focuses on preventing prompt injection, tool misuse, and unsafe behavior in AI agents, ensuring they operate safely and reliably. It helps identify agent risks, protect against evolving attack vectors like data exfiltration, and align AI agents with security standards for compliance. TrojAI is enterprise-proven, customizable, scalable, and offers flexible deployment options, including self-hosting, to meet stringent security needs.

Maibrain is an innovative AI platform designed to preserve knowledge and memories by transforming personal information into intelligent digital agents. The platform allows users to store photos, videos, texts, and audio, creating a comprehensive digital legacy. A key feature is voice cloning, which enables interactive and personalized chat experiences with the digital agents, simulating conversations with loved ones. Maibrain aims to automate processes, enhance customer service, and empower businesses through its advanced technology. It offers both a Free plan with limited features and a Premium plan that includes advanced capabilities like unlimited interactive chat, custom chat links, avatar creation, and scheduling appointments, making it suitable for individuals and businesses looking to digitize and interact with their preserved knowledge.

bindsnet is a Python package designed for simulating spiking neural networks (SNNs) on both CPUs and GPUs, leveraging PyTorch's tensor functionality. It is specifically geared towards the development of biologically inspired algorithms for machine learning, making it a valuable tool for researchers. The package facilitates ongoing research in applying SNNs to machine learning (ML) and reinforcement learning (RL) problems. It allows users to convert ordinary differential equations (ODEs) describing neuron dynamics into difference equations for approximation, utilizing PyTorch's powerful `torch.Tensor` objects and `torch.nn.functional` submodule. This enables the creation of SNN architectures with features like convolution or pooling functions, and supports spike-timing-dependent plasticity (STDP) for weight modification.