Computer Vision

Designed for computer vision tasks, this toolkit simplifies segmentation, object detection, and tracking with integrated annotation and image processing tools.

Status : Beta test

The LabVIEW Computer Vision Module provides a robust suite of tools designed to handle and process visual data effectively. This module enables you to develop and deploy advanced computer vision applications that can perform real-time image processing, object detection, pattern recognition, and automated inspection within the LabVIEW environment.

Deep Learning

A comprehensive solution for training, optimizing, and deploying deep learning models, integrating ONNX Runtime for multi-platform compatibility and enhanced performance.

Status : Beta test

The LabVIEW Deep Learning Module is a comprehensive framework designed to streamline your deep learning projects. It allows you to load, edit, design, train, execute, integrate, and deploy any deep learning model with efficiency. Whether you’re working on computer vision, predictive analytics, or any other AI-driven application, this module provides the tools you need to build sophisticated neural network models directly within the familiar LabVIEW environment.

Accelerator

An optimized environment for leveraging GPU power, enabling accelerated inference and model optimization through standards like TensorRT and ONNX.

Status : Beta test

The LabVIEW GPU Processing Module is engineered to enhance your data processing capabilities by utilizing the power of GPUs. This module allows LabVIEW users to accelerate both upstream and downstream data processing tasks through parallel processing, significantly speeding up computations and data throughput.

GenAI

Specifically designed to generate content with generative AI models, such as text and images, seamlessly integrating into industrial and academic workflows.

Status : To be developed

The GenAI Toolkit is a powerful framework designed to streamline your generative AI projects. It allows you to import, fine-tune, and execute Large Language Models (LLMs), Small Language Models (SLMs), and Vision-Language Models (VLMs) with ease. Whether you’re working on predictive analytics, conversational AI, or any other AI-driven application, this module provides the tools you need to seamlessly integrate intelligent agents into your software architecture, all within the familiar LabVIEW environment.

Robotics

A suite of tools for simulation, control, and integration of robotic systems, with advanced applications in vision and planning.

Status : To be developed

The Robotics Toolkit is a powerful framework designed to streamline your robotics projects. It allows you to simulate, control, and deploy robotic systems with ease. Whether you’re working on motion planning, autonomous navigation, or sensor integration, this toolkit provides the tools you need to seamlessly integrate intelligent robotic agents into your hardware and software architecture, all within the familiar LabVIEW environment.

Netron

An intuitive model visualization tool designed to explore and analyze neural network architectures with ease, providing a clear and interactive representation of each layer and its parameters.

Status : Beta test

Netron is an indispensable tool for visualizing and analyzing deep learning models from frameworks like Keras, TensorFlow, PyTorch, and ONNX. It provides an intuitive interface to explore model architectures, inspect layers, and understand parameter structures. With features like detailed layer visualization, parameter inspection, and compatibility with LabVIEW for the Deep Learning Module, Netron simplifies model debugging, optimization, and integration into production workflows. It’s the go-to solution for clear, interactive insights into your AI models.

FIG

An interactive graphical tool based on LabVIEW, designed to simplify model development and real-time visualization.

Status : Beta test

FIG (File Importator and Generator) is an essential toolkit for importing and converting deep learning models from Keras, TensorFlow, PyTorch, and ONNX. It allows model visualization with Netron, generates classic text summaries, and imports models into LabVIEW for use with the Deep Learning Module. It offers functionalities like model conversion, quantization, LabVIEW integration, visualization, and exportation.

LMS

An integrated learning platform to track, manage, and enhance training processes in AI and new technologies.

Status : To be developed

Language Model Studio (LMS) is a versatile tool designed for importing, fine-tuning, executing, and deploying any LLM (Large Language Model) or SLM (Small Language Model). It also offers model quantization capabilities and allows the creation of chat agents and Retrieval-Augmented Generation (RAG) systems.

Annotator

A modular creation environment to assemble, optimize, and test complex model architectures without coding.

Status : In development

Image Annotator is a comprehensive toolkit designed for importing, editing, and annotating any dataset. It offers advanced preprocessing and data augmentation capabilities. Additionally, the tool enables the generation of Deep Learning projects for training and executing object detection or segmentation models, providing a wide selection of models and deep learning training architectures.

Builder

An intuitive tool for annotating and preparing datasets, essential for creating accurate and robust models.

Status : To be developed

The LabVIEW Executable Builder is designed to streamline deployment for your projects using the SOTA toolkits. It efficiently builds, packages, and deploys executable applications, importing necessary DLLs for smooth functionality. Ideal for complex software, high-performance computing, or any robust deployment needs, this tool provides everything to create and distribute sophisticated executables.

Data

A centralized solution for managing, processing, and analyzing large datasets, tailored to various use cases.

Status : To be developed

The Data Toolkit is a robust solution for managing, processing, and transforming datasets for AI workflows. It simplifies the entire data preparation lifecycle, from cleaning and normalization to augmentation and formatting, ensuring that your datasets are optimized for training and analysis. With seamless integration into cloud storage and database systems, the Data Toolkit empowers you to maintain consistency and scalability across large-scale AI projects while providing advanced features like version control and automated preprocessing pipelines.

Super Mario

A playful environment based on the iconic game, used for research in reinforcement learning and artificial intelligence.

Status : Coming soon

An OpenAI Gym environment for Super Mario Bros. and Super Mario Bros. 2 (Lost Levels) on the NES enabling full control over the environment. This setup includes all 32 stages and ensures efficient training by only sending reward-able gameplay frames to agents, excluding cut-scenes and loading screens.

ViZDoom

An immersive platform for testing AI algorithms in first-person shooter simulation environments.

Status : Coming soon

ViZDoom is an environment allowing the development of AI bots that play DOOM using visual information from the screen buffer. Primarily intended for research in machine visual learning and deep reinforcement learning, ViZDoom supports various learning approaches, including reinforcement learning, learning from demonstration, and apprenticeship learning.

Gymnasium

A modern extension for OpenAI Gym, offering a wide variety of reinforcement learning environments.

Status : Available SOTA beta

Gymnasium is now available in LabVIEW, providing an API for all single-agent reinforcement learning environments. This project includes implementations of common environments such as CartPole, Pendulum, MountainCar, MuJoCo, and more, making it a comprehensive platform for developing and testing AI algorithms in a variety of settings.

Atari

A collection of iconic games used to train and evaluate machine learning models.

Status : Coming soon

Atari environments are now available in LabVIEW via the OpenAI Gym, simulated through the Arcade Learning Environment (ALE) using the Stella emulator. These environments offer a diverse set of classic Atari games, facilitating research in reinforcement learning and providing a robust platform for developing and testing AI algorithms.

Highway

A realistic simulator for developing AI algorithms applied to autonomous vehicles and mobility.

Status : Coming soon

The Highway environment is now available in LabVIEW, offering a collection of environments designed for autonomous driving and tactical decision-making tasks. This setup provides a versatile platform for developing and testing AI algorithms in realistic driving scenarios, facilitating research and innovation in autonomous vehicle technologies.

Stable Retro

A retro-optimized environment for exploring AI models on classic games and systems, focusing on stability and reproducibility.

Status : Coming soon

Stable Retro, now available in LabVIEW, is a fork of gym-retro that lets you turn classic video games for reinforcement learning. It includes additional games, emulators, and supported platforms. Unlike gym-retro, which is now in maintenance mode, Stable Retro accepts new games, platforms, and bug fixes, allowing for continuous updates and improvements through pull requests.