Open Source Deep Learning Frameworks - Page 2
Sort By:
Deep Learning Frameworks
View 108 business solutions-
MongoDB Atlas runs apps anywhereMongoDB Atlas gives you the freedom to build and run modern applications anywhere—across AWS, Azure, and Google Cloud. With global availability in over 115 regions, Atlas lets you deploy close to your users, meet compliance needs, and scale with confidence across any geography.
-
Go From AI Idea to AI App FastAccess Gemini 3 and 200+ models. Build chatbots, agents, or custom models with built-in monitoring and scaling.
-
1
Torch-TensorRT
PyTorch/TorchScript/FX compiler for NVIDIA GPUs using TensorRT
Torch-TensorRT is a compiler for PyTorch/TorchScript, targeting NVIDIA GPUs via NVIDIA’s TensorRT Deep Learning Optimizer and Runtime. Unlike PyTorch’s Just-In-Time (JIT) compiler, Torch-TensorRT is an Ahead-of-Time (AOT) compiler, meaning that before you deploy your TorchScript code, you go through an explicit compile step to convert a standard TorchScript program into a module targeting a TensorRT engine. Torch-TensorRT operates as a PyTorch extension and compiles modules that integrate into the JIT runtime seamlessly. After compilation using the optimized graph should feel no different than running a TorchScript module. You also have access to TensorRT’s suite of configurations at compile time, so you are able to specify operating precision (FP32/FP16/INT8) and other settings for your module. -
2
Apache MXNet (incubating)
A flexible and efficient library for deep learning
Apache MXNet is an open source deep learning framework designed for efficient and flexible research prototyping and production. It contains a dynamic dependency scheduler that automatically parallelizes both symbolic and imperative operations. On top of this is a graph optimization layer, overall making MXNet highly efficient yet still portable, lightweight and scalable. -
3
D2L.ai
Interactive deep learning book with multi-framework code
Interactive deep learning book with multi-framework code, math, and discussions. Adopted at 300 universities from 55 countries including Stanford, MIT, Harvard, and Cambridge. This open-source book represents our attempt to make deep learning approachable, teaching you the concepts, the context, and the code. The entire book is drafted in Jupyter notebooks, seamlessly integrating exposition figures, math, and interactive examples with self-contained code. Offers sufficient technical depth to provide a starting point on the path to actually becoming an applied machine learning scientist. -
4
Ludwig
A codeless platform to train and test deep learning models
Ludwig is a toolbox built on top of TensorFlow that allows to train and test deep learning models without the need to write code. All you need to provide is a CSV file containing your data, a list of columns to use as inputs, and a list of columns to use as outputs, Ludwig will do the rest. Simple commands can be used to train models both locally and in a distributed way, and to use them to predict on new data. -
Custom VMs From 1 to 96 vCPUs With 99.95% UptimeLive migration and automatic failover keep workloads online through maintenance. One free e2-micro VM every month.
-
5
Lux.jl
Elegant and Performant Deep Learning
Lux.jl is a lightweight and extensible deep learning framework in Julia designed for speed, composability, and clarity. Unlike traditional machine learning libraries that bundle training logic and models, Lux separates model definitions from training routines, encouraging modularity and ease of experimentation. It integrates seamlessly with SciML and other Julia packages, supporting neural differential equations and scientific machine learning workflows. -
6
ONNX
Open standard for machine learning interoperability
ONNX is an open format built to represent machine learning models. ONNX defines a common set of operators - the building blocks of machine learning and deep learning models - and a common file format to enable AI developers to use models with a variety of frameworks, tools, runtimes, and compilers. Open Neural Network Exchange (ONNX) is an open ecosystem that empowers AI developers to choose the right tools as their project evolves. ONNX provides an open source format for AI models, both deep learning and traditional ML. It defines an extensible computation graph model, as well as definitions of built-in operators and standard data types. Currently we focus on the capabilities needed for inferencing (scoring). ONNX is widely supported and can be found in many frameworks, tools, and hardware. Enabling interoperability between different frameworks and streamlining the path from research to production helps increase the speed of innovation in the AI community. -
7
PyTorch Geometric
Geometric deep learning extension library for PyTorch
It consists of various methods for deep learning on graphs and other irregular structures, also known as geometric deep learning, from a variety of published papers. In addition, it consists of an easy-to-use mini-batch loader for many small and single giant graphs, a large number of common benchmark datasets (based on simple interfaces to create your own), and helpful transforms, both for learning on arbitrary graphs as well as on 3D meshes or point clouds. We have outsourced a lot of functionality of PyTorch Geometric to other packages, which needs to be additionally installed. These packages come with their own CPU and GPU kernel implementations based on C++/CUDA extensions. We do not recommend installation as root user on your system python. Please setup an Anaconda/Miniconda environment or create a Docker image. We provide pip wheels for all major OS/PyTorch/CUDA combinations. -
8
Python Outlier Detection
A Python toolbox for scalable outlier detection
PyOD is a comprehensive and scalable Python toolkit for detecting outlying objects in multivariate data. This exciting yet challenging field is commonly referred as outlier detection or anomaly detection. PyOD includes more than 30 detection algorithms, from classical LOF (SIGMOD 2000) to the latest COPOD (ICDM 2020) and SUOD (MLSys 2021). Since 2017, PyOD [AZNL19] has been successfully used in numerous academic researches and commercial products [AZHC+21, AZNHL19]. PyOD has multiple neural network-based models, e.g., AutoEncoders, which are implemented in both PyTorch and Tensorflow. PyOD contains multiple models that also exist in scikit-learn. It is possible to train and predict with a large number of detection models in PyOD by leveraging SUOD framework. A benchmark is supplied for select algorithms to provide an overview of the implemented models. In total, 17 benchmark datasets are used for comparison, which can be downloaded at ODDS. -
9
Requests for Research
A living collection of deep learning problems
Requests for Research is an OpenAI repository that collects and organizes open research ideas in artificial intelligence. It is structured as a curated list of project proposals, challenges, and exploratory directions suggested by OpenAI researchers for the broader community. Each request highlights a specific problem area, often with context, motivation, and possible approaches, serving as inspiration for independent researchers, students, and practitioners. The repository is intended to foster collaboration and accelerate progress by sharing promising questions that OpenAI itself may not have the resources to fully pursue. While the ideas are not guaranteed to be unique or unexplored, they reflect areas OpenAI believes would benefit from more investigation. The project functions as a living document of research directions rather than a codebase, making it an accessible entry point for those looking to contribute to the advancement of AI. -
Forever Free Full-Stack Observability | Grafana CloudBuilt on open standards like Prometheus and OpenTelemetry, Grafana Cloud includes Kubernetes Monitoring, Application Observability, Incident Response, plus the AI-powered Grafana Assistant. Get started with our generous free tier today.
-
10
DeepDetect
Deep Learning API and Server in C++14 support for Caffe, PyTorch
The core idea is to remove the error sources and difficulties of Deep Learning applications by providing a safe haven of commoditized practices, all available as a single core. While the Open Source Deep Learning Server is the core element, with REST API, and multi-platform support that allows training & inference everywhere, the Deep Learning Platform allows higher level management for training neural network models and using them as if they were simple code snippets. Ready for applications of image tagging, object detection, segmentation, OCR, Audio, Video, Text classification, CSV for tabular data and time series. Neural network templates for the most effective architectures for GPU, CPU, and Embedded devices. Training in a few hours and with small data thanks to 25+ pre-trained models. Full Open Source, with an ecosystem of tools (API clients, video, annotation, ...) Fast Server written in pure C++, a single codebase for Cloud, Desktop & Embedded. -
11
DeepSpeed MII
MII makes low-latency and high-throughput inference possible
MII makes low-latency and high-throughput inference possible, powered by DeepSpeed. The Deep Learning (DL) open-source community has seen tremendous growth in the last few months. Incredibly powerful text generation models such as the Bloom 176B, or image generation model such as Stable Diffusion are now available to anyone with access to a handful or even a single GPU through platforms such as Hugging Face. While open-sourcing has democratized access to AI capabilities, their application is still restricted by two critical factors: inference latency and cost. DeepSpeed-MII is a new open-source python library from DeepSpeed, aimed towards making low-latency, low-cost inference of powerful models not only feasible but also easily accessible. MII offers access to the highly optimized implementation of thousands of widely used DL models. MII-supported models achieve significantly lower latency and cost compared to their original implementation. -
12
Deeplearning.ai
Study notes, summaries, and auxiliary materials for deep learning
Deeplearning.ai collects study notes, summaries, and auxiliary materials aligned with the popular deep learning course series many learners take early in their AI journey. It distills core ideas such as optimization, regularization, convolutional networks, sequence models, and practical training tricks. The explanations aim to bridge theory and practice, often connecting mathematical intuition to code-level implications. By organizing the content as “books” or structured notes, it gives students a consistent reference to revisit as models and tooling evolve. Many learners use it to supplement course videos, reinforcing concepts before implementing assignments or projects. As a consolidated guide, it reduces context-switching and helps build a durable mental model of deep learning fundamentals. -
13
Detectron2
Next-generation platform for object detection and segmentation
Detectron2 is Facebook AI Research's next generation software system that implements state-of-the-art object detection algorithms. It is a ground-up rewrite of the previous version, Detectron, and it originates from maskrcnn-benchmark. It is powered by the PyTorch deep learning framework. Includes more features such as panoptic segmentation, Densepose, Cascade R-CNN, rotated bounding boxes, PointRend, DeepLab, etc. Can be used as a library to support different projects on top of it. We'll open source more research projects in this way. It trains much faster. Models can be exported to TorchScript format or Caffe2 format for deployment. With a new, more modular design, Detectron2 is flexible and extensible, and able to provide fast training on single or multiple GPU servers. Detectron2 includes high-quality implementations of state-of-the-art object detection. -
14
Exposure Correction
Learning multi-scale deep model correcting over- and under- exposed
Exposure_Correction is a research project that provides the implementation for the paper Learning Multi-Scale Photo Exposure Correction (CVPR 2021). The repository focuses on correcting poorly exposed photographs, handling both underexposure and overexposure using a deep learning approach. The method employs a multi-scale framework that learns to enhance images by adjusting exposure levels across different spatial resolutions. This allows the model to preserve fine details while correcting global lighting inconsistencies. The repository includes pre-trained models, datasets, and training/testing code to enable reproducibility and experimentation. By leveraging this framework, researchers and developers can apply exposure correction to a wide range of natural images, improving visual quality without manual editing. The project serves both as a research reference and a practical tool for computational photography and image enhancement. -
15
Horovod
Distributed training framework for TensorFlow, Keras, PyTorch, etc.
Horovod was originally developed by Uber to make distributed deep learning fast and easy to use, bringing model training time down from days and weeks to hours and minutes. With Horovod, an existing training script can be scaled up to run on hundreds of GPUs in just a few lines of Python code. Horovod can be installed on-premise or run out-of-the-box in cloud platforms, including AWS, Azure, and Databricks. Horovod can additionally run on top of Apache Spark, making it possible to unify data processing and model training into a single pipeline. Once Horovod has been configured, the same infrastructure can be used to train models with any framework, making it easy to switch between TensorFlow, PyTorch, MXNet, and future frameworks as machine learning tech stacks continue to evolve. Start scaling your model training with just a few lines of Python code. Scale up to hundreds of GPUs with upwards of 90% scaling efficiency. -
16
MATLAB Deep Learning Model Hub
Discover pretrained models for deep learning in MATLAB
Discover pre-trained models for deep learning in MATLAB. Pretrained image classification networks have already learned to extract powerful and informative features from natural images. Use them as a starting point to learn a new task using transfer learning. Inputs are RGB images, the output is the predicted label and score. -
17
MONAI
AI Toolkit for Healthcare Imaging
The MONAI framework is the open-source foundation being created by Project MONAI. MONAI is a freely available, community-supported, PyTorch-based framework for deep learning in healthcare imaging. It provides domain-optimized foundational capabilities for developing healthcare imaging training workflows in a native PyTorch paradigm. Project MONAI also includes MONAI Label, an intelligent open source image labeling and learning tool that helps researchers and clinicians collaborate, create annotated datasets, and build AI models in a standardized MONAI paradigm. MONAI is an open-source project. It is built on top of PyTorch and is released under the Apache 2.0 license. Aiming to capture best practices of AI development for healthcare researchers, with an immediate focus on medical imaging. Providing user-comprehensible error messages and easy to program API interfaces. Provides reproducibility of research experiments for comparisons against state-of-the-art implementations. -
18
Metaflow
A framework for real-life data science
Metaflow is a human-friendly Python library that helps scientists and engineers build and manage real-life data science projects. Metaflow was originally developed at Netflix to boost productivity of data scientists who work on a wide variety of projects from classical statistics to state-of-the-art deep learning. -
19
Robust Tube MPC
Example implementation for robust model predictive control using tube
robust-tube-mpc is a MATLAB implementation of robust tube-based Model Predictive Control (MPC). The framework provides tools to design and simulate controllers that maintain stability and constraint satisfaction in the presence of bounded disturbances. Tube-based MPC achieves robustness by combining a nominal trajectory planner with an error feedback controller that keeps the actual system state within a "tube" around the nominal trajectory. This repository includes example scripts and implementations demonstrating how to apply the method to control problems. It is particularly useful for researchers, students, and engineers exploring robust control strategies in uncertain environments. By offering a structured implementation, robust-tube-mpc makes it easier to study and extend advanced MPC techniques for real-world applications. -
20
TTS
Deep learning for text to speech
TTS is a library for advanced Text-to-Speech generation. It's built on the latest research, was designed to achieve the best trade-off among ease-of-training, speed, and quality. TTS comes with pre-trained models, tools for measuring dataset quality, and is already used in 20+ languages for products and research projects. Released models in PyTorch, Tensorflow and TFLite. Tools to curate Text2Speech datasets underdataset_analysis. Demo server for model testing. Notebooks for extensive model benchmarking. Modular (but not too much) code base enabling easy testing for new ideas. Text2Spec models (Tacotron, Tacotron2, Glow-TTS, SpeedySpeech). Speaker Encoder to compute speaker embeddings efficiently. Vocoder models (MelGAN, Multiband-MelGAN, GAN-TTS, ParallelWaveGAN, WaveGrad, WaveRNN). If you are only interested in synthesizing speech with the released TTS models, installing from PyPI is the easiest option. -
21
Thinc
A refreshing functional take on deep learning
Thinc is a lightweight deep learning library that offers an elegant, type-checked, functional-programming API for composing models, with support for layers defined in other frameworks such as PyTorch, TensorFlow and MXNet. You can use Thinc as an interface layer, a standalone toolkit or a flexible way to develop new models. Previous versions of Thinc have been running quietly in production in thousands of companies, via both spaCy and Prodigy. We wrote the new version to let users compose, configure and deploy custom models built with their favorite framework. Switch between PyTorch, TensorFlow and MXNet models without changing your application, or even create mutant hybrids using zero-copy array interchange. Develop faster and catch bugs sooner with sophisticated type checking. Trying to pass a 1-dimensional array into a model that expects two dimensions? That’s a type error. Your editor can pick it up as the code leaves your fingers. -
22
Triton Inference Server
The Triton Inference Server provides an optimized cloud
Triton Inference Server is an open-source inference serving software that streamlines AI inferencing. Triton enables teams to deploy any AI model from multiple deep learning and machine learning frameworks, including TensorRT, TensorFlow, PyTorch, ONNX, OpenVINO, Python, RAPIDS FIL, and more. Triton supports inference across cloud, data center, edge, and embedded devices on NVIDIA GPUs, x86 and ARM CPU, or AWS Inferentia. Triton delivers optimized performance for many query types, including real-time, batched, ensembles, and audio/video streaming. Provides Backend API that allows adding custom backends and pre/post-processing operations. Model pipelines using Ensembling or Business Logic Scripting (BLS). HTTP/REST and GRPC inference protocols based on the community-developed KServe protocol. A C API and Java API allow Triton to link directly into your application for edge and other in-process use cases. -
23
Consistent Depth
We estimate dense, flicker-free, geometrically consistent depth
Consistent Depth is a research project developed by Facebook Research that presents an algorithm for reconstructing dense and geometrically consistent depth information for all pixels in a monocular video. The system builds upon traditional structure-from-motion (SfM) techniques to provide geometric constraints while integrating a convolutional neural network trained for single-image depth estimation. During inference, the model fine-tunes itself to align with the geometric constraints of a specific input video, ensuring stable and realistic depth maps even in less-constrained regions. This approach achieves improved geometric consistency and visual stability compared to prior monocular reconstruction methods. The project can process challenging hand-held video footage, including those with moderate dynamic motion, making it practical for real-world usage. -
24
DeepLearning
Deep Learning (Flower Book) mathematical derivation
" Deep Learning " is the only comprehensive book in the field of deep learning. The full name is also called the Deep Learning AI Bible (Deep Learning) . It is edited by three world-renowned experts, Ian Goodfellow, Yoshua Bengio, and Aaron Courville. Includes linear algebra, probability theory, information theory, numerical optimization, and related content in machine learning. At the same time, it also introduces deep learning techniques used by practitioners in the industry, including deep feedforward networks, regularization, optimization algorithms, convolutional networks, sequence modeling and practical methods, and investigates topics such as natural language processing, Applications in speech recognition, computer vision, online recommender systems, bioinformatics, and video games. Finally, the Deep Learning book provides research directions covering theoretical topics including linear factor models, autoencoders, representation learning, structured probabilistic models, etc. -
25
End-to-End Negotiator
Deal or No Deal? End-to-End Learning for Negotiation Dialogues
End-to-End Negotiator is a PyTorch-based research framework developed by Facebook AI Research to train neural agents capable of conducting strategic negotiations in natural language. The project implements the models presented in two key papers: “Deal or No Deal? End-to-End Learning for Negotiation Dialogues” and “Hierarchical Text Generation and Planning for Strategic Dialogue”. It enables agents to plan, reason, and communicate effectively to maximize outcomes in multi-turn negotiations over shared resources. The framework provides code for both supervised learning (training from human dialogue data) and reinforcement learning (via self-play and rollout-based planning). It introduces a hierarchical latent model, where high-level intents are first clustered and then translated into coherent language, improving dialogue diversity and goal consistency. The repository also includes the Negotiate dataset, comprising over 5,800 dialogues across 2,200 unique scenarios.
