Tame the physics of your
projects in hours!

SIML.ai is a software platform for working with
high-performance AI-based numerical simulators.

simulation-studio-preview.jpeg

Physics & AI

SIML.ai combines state-of-the-art machine learning with physics simulation.

Fast

AI-based simulators responds to your input in real-time, which means interactive visualization.

Web-based

No need to install anything or be limited to specific platforms. Everything happens through web interface.

Painless scalability

Our platform deals with the complexities of setting up cloud or HPC infrastructure for you.

Vast majority of technologies in the world come to life through months or years of extensive simulation during their development.

High-performance computing, parallel processing and GPUs helped push the computation time from months to weeks. With the help of applied machine learning, we are seeing a reduction from weeks to days. We think that’s not enough.

At DimensionLab, we are building tools for engineers and researchers to tame the physics of their projects in hours. Collectively, they make up a cohesive platform we call SIML.ai.

Under the hood, SIML.ai consists of two parts - Model Engineer and Simulation Studio.

Screenshot 2022-08-24 at 10.48.06.png

Model Engineer

Train and optimize extremely fast physics simulators using deep learning techniques through web-based Model Engineer application.

Standing on the shoulders of giants

Leveraging powerful Modulus neural network framework by NVIDIA, Model Engineer provides a high-level layer on top which makes it super easy to build parameterized surrogate models with near-real-time latency.

Datasets management

Construct large datasets from classical simulation exports or physical sensors that collect precise measurements from real-world experiments.

Building blocks

Quickly develop the model architecture for the simulators’ desired capabilities and physical constraints. Customizable to the bone through code editor.

High performance computing, automated

One-click A100 GPUs. Train and aggressively optimize learnable simulators in high-performance, GPU-powered cloud or HPC centers without the need to deal with the complexities of managing the cloud infrastructure.

Alpha version available for early adopters! We're giving access on rolling basis through our waitlist.

Simulation Studio

Hybrid web and native application for solving engineering and scientific problems leveraging pre-trained and optimized learned simulators.

Superfast

Simulations are computed by inferring trained neural network models, achieving speedups of 1,000-100,000x compared to classical simulation software running on GPUs.

Interactive "in-situ" visualization

The time it takes to compute one timestep even of highly irregular simulation domain is in low tens of milliseconds, resulting in real-time visualization of simulating physical phenomena.

Unreal rendering performance

High-fidelity visualization rendering is achieved by leveraging the powerful Unreal Engine under the hood.

Virtual reality

Immerse yourself into the running simulation and gain detailed insights by slicing through geometries with your hands or instantly see measurements by pointing your fingers.

Expected release: Q4 2022
simulation-studio-preview.jpeg

We believe in the democratization of scientific-grade simulation tools by making it easy for anyone to develop physics-based simulations and deploy them in their workflows, regardless of their technical skills.

Be the first to try the platform!

Help us test the product, send feedback and get free access for a month after we launch!

Interested in funding us?

Drop us an email and we'll send you our pitchdeck.

climaccelerator-kic-logo.png