Open-source alternatives to Tiny Cuda Nn

TileLang is a Python-embedded domain-specific language compiler that JIT-compiles and autotunes GPU kernels. It uses a tile-based DSL, automatic software pipelining, and parallel autotuning to generate optimized GPU kernels at runtime. It supports tensor core operations with Pythonic syntax, automatic memory management, and thread mapping. The compiler searches over tile sizes, thread counts, and scheduling policies, compiling and benchmarking candidates in parallel to find the fastest kernel. It also caches compiled binaries and tuning results to disk for reuse across sessions. TileLang inc

This project is a static educational website and comprehensive curriculum focused on computer vision and deep learning. It serves as a public repository of instructional materials, lecture notes, and technical guides specifically detailing convolutional neural networks and visual recognition. The site is developed using static-site generation to host course documentation and student project directories. It provides structured academic resources that guide learners through image classification, generative modeling, and the implementation of various neural network architectures. The curriculum

This project is a machine learning educational archive and technical documentation collection. It serves as a deep learning tutorial series and implementation guide, providing theoretical explanations and practical walkthroughs for constructing and optimizing neural networks. The content focuses on the design and construction of diverse model architectures, including convolutional neural networks, Long Short-Term Memory networks, and generative adversarial networks. It details specific implementation patterns for autoencoders, sentiment analysis models, and various classification approaches.

Instant-ngp is a high-performance neural graphics engine and toolkit designed for 3D reconstruction and the rendering of neural radiance fields. It provides an integrated framework for generating photorealistic volumetric representations from sets of two-dimensional images by optimizing continuous neural scene models. The project distinguishes itself through a focus on rapid training and real-time inference, achieved by mapping spatial coordinates into compact feature grids. By utilizing multiresolution hash encoding and fused processing kernels, the system minimizes computational overhead an

This project is an educational codebase and reference library that translates theoretical deep learning concepts into executable PyTorch code. It serves as a practical implementation of a deep learning textbook, providing a course-like structure of guided exercises and architectural examples for learning purposes. The repository includes a library of standard neural network architectures, including linear, convolutional, recurrent, and transformer models. It specifically implements a variety of deep learning patterns such as multilayer perceptrons, VGG networks, gated recurrent units, and lon

This project is an educational resource providing practical code examples and implementations of machine learning algorithms using the Python language. It serves as a guide for constructing predictive pipelines, clustering models, and dimensionality reduction within the Scikit-Learn ecosystem. The repository includes comprehensive demonstrations for supervised and unsupervised learning, as well as detailed examples for implementing neural networks and deep architectures. It also provides practical guidance on exporting model parameters to JSON and wrapping trained models in web APIs for produ

mmagic is a multimodal training pipeline and framework for generative AI, focusing on visual synthesis and restoration. It provides the infrastructure to build and train models for tasks such as text-to-image and text-to-video generation, 3D-aware content synthesis, and high-fidelity image translation using diffusion models and generative adversarial networks. The project distinguishes itself through specialized capabilities for generative model personalization, including techniques for fine-tuning subjects and styles. It also supports advanced visual manipulations such as latent space interp

Gaussian Splatting is a computational framework designed to transform sparse sets of two-dimensional photographs into photorealistic, interactive three-dimensional scene representations. The system functions as a reconstruction tool and rendering engine, enabling the conversion of image data into volumetric models that support novel view synthesis. The project represents scenes as a collection of anisotropic three-dimensional Gaussians, which store position, opacity, color, and covariance data. It distinguishes itself through a differentiable tile-based rasterization process that projects the

This project is an educational resource and comprehensive guide for implementing and deploying deep learning models using the PyTorch framework. It provides a structured learning curriculum consisting of tutorials and notebooks that cover neural network architectures, data pipelines, and model optimization across multiple AI domains. The curriculum includes practical implementation guides for building convolutional networks, transformers, and recurrent models. It specifically focuses on workflows for computer vision, including image classification, object detection, and segmentation, as well

This repository is a comprehensive educational program and deep learning framework designed to teach practical deep learning using PyTorch through notebooks and code examples. It serves as a high-level library for building, training, and deploying neural networks, acting as a model training orchestrator that coordinates PyTorch models, optimizers, and loss functions. The project provides specialized toolkits for computer vision, natural language processing, and tabular data preprocessing. It distinguishes itself through advanced training controls such as discriminative learning rates, a two-w

Composer is a PyTorch distributed training framework designed for scaling large-scale models across multi-node GPU clusters. It functions as a large language model trainer, a distributed model optimizer, and a training lifecycle manager. The project differentiates itself as a deep learning regularization library, providing specialized optimization techniques such as Sharpness Aware Minimization, MixUp, and CutMix to improve model generalization. It further distinguishes its training flow through the use of sequence length warmup, progressive layer freezing, and sharded-state checkpointing for

ConvNetJS is a JavaScript deep learning library and neural network training engine designed for client-side machine learning. It functions as a framework for building, training, and running convolutional neural networks directly within a web browser without the need for a backend server. The library specializes in image recognition and pattern analysis using convolutional and pooling layers. It enables the creation of models for classification and regression tasks, as well as the development of reinforcement learning agents that optimize behavior through trial and error in simulated environme

SD.Next is an all-in-one web interface and multi-backend inference engine for generating, editing, and processing images and videos using diffusion models. It functions as a comprehensive tool for diffusion model management and an automated image processing pipeline for bulk operations. The project is distinguished by its hardware-backend abstraction layer, which provides automatic detection and acceleration for NVIDIA CUDA, AMD ROCm, Intel OpenVINO, and DirectML. It features a headless generative API and a programmatic command interface, allowing users to trigger tasks via REST API or CLI wi

Neural Enhance is a deep learning image upscaler and restoration tool designed to increase image resolution and remove blur. It functions as a neural image restoration utility for eliminating noise and JPEG artifacts, and includes a framework for training and tuning custom neural network models against image datasets. The system utilizes a containerized environment to offload tensor calculations to GPU cores, speeding up neural network inference. It features a batch processing pipeline that queues multiple image files in sequence to maximize hardware throughput. Capabilities include domain-s

Swift-AI is an on-device library for training and running fully-connected neural networks on Apple platforms. It is designed to enable machine learning workflows directly on mobile hardware without requiring cloud connectivity, supporting both training and inference for tasks such as image classification and handwriting recognition. The library distinguishes itself through deep integration with Apple's ecosystem, leveraging Metal for GPU-accelerated matrix operations and using Swift's compiler infrastructure to parse a domain-specific language for neural network definitions. It implements rev

Caffe is a high-performance deep learning framework and convolutional neural network library designed for training and deploying neural networks. It functions as a GPU-accelerated machine learning engine with a core implemented in C++ to enable high-throughput tensor operations. The project utilizes a declarative configuration system where model architectures and hyperparameters are defined in external text files, separating the network design from the execution code. It includes a model serialization system to export trained weights and topologies into binary files for efficient deployment a

Map-anything is a 3D scene reconstruction framework and neural geometry estimator designed to transform two-dimensional images into metric three-dimensional spatial representations using feed-forward neural networks. It provides a specialized toolkit for predicting camera intrinsics and ray directions from single images without requiring external geometric metadata. The project includes a 3D model benchmarking suite that utilizes a unified model wrapper to standardize outputs from diverse reconstruction models. This allows for consistent evaluation and accuracy measurement across various spat

This project is a deep learning library and neural network training framework built for the TensorFlow ecosystem. It functions as a structured repository of algorithms and tools designed to execute iterative learning routines, fit complex datasets to predictive models, and manage the deployment of trained neural networks. The library provides a standardized interface for machine learning research prototyping, allowing users to experiment with various architectures and validate data models. It supports the full lifecycle of model development, from the initial training of neural networks on cus

SAM 3D Objects is a promptable foundation model that recovers 3D objects and human meshes from single images. It converts masked objects in a single photograph into full 3D models with pose, shape, texture, and layout, while also producing complete 3D human body meshes from the same input. The system integrates promptable segmentation to isolate objects and humans before reconstruction, then aligns the independently reconstructed 3D elements into a shared coordinate space. This enables scene-level understanding where multiple 3D reconstructions from the same image coexist in a common coordina

Darknet is a high-performance C-based inference engine and computer vision library designed for real-time object identification and localization. It serves as a neural network framework for training and deploying detection models using the YOLO architecture, providing a toolset for deep learning training and deployment. The project differentiates itself through a C and CUDA implementation that enables hardware acceleration for matrix multiplication and inference speed optimization. It provides a shared library interface for embedding detection capabilities into external applications and suppo

Tensorpack is a high-performance TensorFlow training framework and distributed deep learning toolkit. It provides a suite of tools for building and training neural networks with a focus on execution speed and architectural flexibility. The project serves as a neural network optimization suite, implementing high-efficiency execution patterns to reduce training overhead. It functions as a parallel data loading pipeline, using automated parallelization to maximize throughput when processing large datasets. The toolkit covers distributed training across multiple GPUs and compute clusters using d

Tensorpack is a high-level TensorFlow neural network framework and research library designed for building and training deep learning models. It provides a collection of reproducible neural network architectures for computer vision, generative tasks, reinforcement learning, and natural language processing. The project distinguishes itself through a specialized deep learning data pipeline that uses pure Python for parallel data loading and streaming. It includes a multi-GPU training orchestrator for distributing workloads via data-parallel strategies and a dedicated interpretability toolkit for

Knowledge-Distillation-Zoo is a framework for neural network model compression that facilitates the transfer of learned patterns from large teacher models to smaller student architectures. It provides a modular environment for executing training pipelines designed to reduce the computational requirements of deep learning models while maintaining predictive accuracy. The library implements knowledge transfer through both logit-based mimicry and feature-map alignment, allowing students to replicate the classification behavior and internal representations of a teacher. It supports teacher-studen

Skorch is a deep learning workflow manager and tensor-based model interface. It provides a consistent API for training and predicting with neural networks within standard machine learning workflows, acting as a hyperparameter optimizer for finding optimal network configurations. The library specializes in wrapping PyTorch neural networks in a scikit-learn compatible interface. This allows tensor-based models to be used within traditional machine learning pipelines and grid search tools, including the mapping of parameter grids to model configurations. The framework covers training lifecycle

Chainer is an open-source deep learning framework built around define-by-run automatic differentiation, where computation graphs are constructed dynamically during forward execution. This imperative approach allows networks to be built using standard Python control flow, with gradients computed automatically through reverse-mode differentiation on the dynamically recorded graph. The framework supports GPU acceleration through a NumPy-compatible array backend with CUDA and cuDNN support, and provides a pluggable device abstraction that lets users switch between CPU and GPU computation without c

DeepGEMM is a suite of specialized GPU kernels and a just-in-time compiler designed for low-precision matrix operations, Mixture-of-Experts models, and attention processing. It provides a library of high-performance matrix multiplication kernels using FP8 precision to increase compute throughput and reduce memory usage. The project features a JIT CUDA kernel compiler that generates and loads optimized compute kernels at runtime to eliminate the need for manual compilation during installation. It includes specialized implementations for grouped matrix multiplication that process multiple group

DeepEP is a distributed model accelerator and expert-parallel communication library designed to optimize the training and inference of large-scale neural networks. It provides specialized GPU communication kernels and a remote GPU memory interface to facilitate high-throughput data exchange between hardware nodes. The system utilizes dynamic kernel generation to compile optimized GPU kernels during execution, removing the need for separate installation compilation steps. It implements virtual-lane traffic isolation to prevent interference between different data streams and employs routing met