SOTA / GRAIPHIC ECOSYSTEM

SOTA. Integrate AI into LabVIEW with ease.

Discover Graiphic’s ecosystem of LabVIEW tools and toolkits: computer vision, Deep Learning, generative AI and CPU/GPU acceleration. An integrated solution that makes it easier to bring AI into your LabVIEW software architectures, from data preparation to industrial deployment.

Download the SOTA installer SOTA documentation Licences and pricing Watch demos

Windows installer · 64-bit. View All Versions

SOTA manager: selecting the Deep Learning Toolkit for LabVIEW — The manager for your LabVIEW ecosystem

Five years of Graiphic R&D / SOTA

Our achievement: bringing AI natively into LabVIEW.

Five years of development brought modern AI into the LabVIEW workflow. With SOTA, engineers build AI applications using LabVIEW VIs, without having to write their application logic in Python or another external language.

From industrial image datasets to deployed models, the ecosystem connects Deep Learning, Computer Vision, GenAI and hardware acceleration with your software architecture.

01 / Integrate

Keep your LabVIEW architecture.

Add model training, inference and image processing to the graphical workflows your engineering team already uses.

02 / Control

Train locally. Keep your data.

Prepare, train and validate models on your industrial image datasets in your own environment, without sending confidential images to a cloud training service.

03 / Accelerate

Put the hardware to work.

Use CPU/GPU execution for supported workloads. The Accelerator Toolkit brings GPU computation into LabVIEW, with public benchmarks and source VIs to examine the results.

04 / Deliver

An ecosystem from data to deployment.

Combine the four core toolkits with Annotator, FIG, Builder, versioned packages and guides to prepare data, import models and integrate them into your applications.

Explore the four toolkits Discover the research behind SOTA

01 / Get started

From installation to your first VI.

The SOTA manager prepares your environment. The guides explain how to install and use each toolkit. The catalog lets you choose Deep Learning, Computer Vision, Accelerator, GenAI and CUDA packages for your LabVIEW version.

01
Install SOTA
Download the SOTA Windows installer and run it on your development machine.
02
Choose a toolkit
Review the guides and examples before adding the module to your LabVIEW project.
03
Test with your data
Adapt an example to your system’s requirements. Our team can help with architecture and integration.

All downloads SOTA documentation

02 / TOOLKITS

A module for every need.

Each toolkit has its own guide and packages by LabVIEW version. Find files, hardware drivers, models and examples in the SOTA catalog.

Deep Learning Toolkit

Neural network training and inference inside LabVIEW.

Documentation Choose your LabVIEW version

Computer Vision Toolkit

Image acquisition, processing, annotation and inspection.

Documentation Choose your LabVIEW version

Accelerator Toolkit

Accelerate supported LabVIEW computations on the GPU.

Documentation Choose your LabVIEW version View benchmarks

GenAI Toolkit

Generative model and local AI workflows in LabVIEW.

Documentation Choose your LabVIEW version

CUDA Toolkit · companion module for GPU computation.

CUDA guide All downloads

03 / APPLICATIONS

From model to LabVIEW project.

SOTA also includes tools to prepare data, import models and build a usable application.

FIG

Import and convert PyTorch, TensorFlow, Keras and ONNX models for LabVIEW use.

Annotator

Annotate and augment image datasets, then train and test vision models.

Builder

Prepare LabVIEW executables with the libraries and files required by Graiphic toolkits.

Netron

Inspect a model’s architecture, layers and parameters.

Watch the Annotator demo Watch a reinforcement learning environment

04 / SOTA · ACCELERATOR TOOLKIT

LabVIEW GPU acceleration, measured.

Accelerator Toolkit is SOTA’s accelerated computing module. Our public LabVIEW GPU Benchmarks repository compares its GPU execution with CuLab, G2CPU and native LabVIEW CPU execution. The method, results and source VIs let you inspect or reproduce the tests. Gains depend on the hardware and workload.

Read report v1.1 Inspect source VIs View GitHub repository