AuraML, a pioneer in generative AI for physical systems, has announced the launch of AuraSim, its Multimodal World Simulation Model. The platform generates physics-ready, high-fidelity simulation environments for robots using inputs such as text descriptions, floorplans, videos, depth maps and point clouds.

Built leveraging NVIDIA Omniverse and NVIDIA Cosmos, AuraML positions itself as a critical infrastructure layer for the global robotics industry, reducing the time and cost required to develop robust physical AI systems.

To support adoption and skill development, AuraML, in collaboration with NVIDIA Inception, is launching The Global Physical AI & Robotics Cohort, a dedicated training programme focused on its World Simulation Model.

Training autonomous systems requires large-scale, high-quality simulation data. AuraML addresses this need by automatically transforming multimodal inputs into ‘sim-ready’ digital twins that adhere to the laws of physics.

Key features of the AuraML World Model include:

Multimodal flexibility: Generates complex 3D environments from 2D floorplans and converts text descriptions into raw sensor data.

Physics-ready outputs: Produces environments with accurate collision boundaries and physical properties, ready for deployment in high-fidelity engines.

Agentic automation: Integrates an automated simulation pipeline powered by AI agents, capable of running thousands of scenario variations without human intervention.

Rapid iteration: Enables developers to stress-test physical AI by generating edge-case scenarios through simple text prompt modifications.

AuraML enables the generation of an entire industrial facility from a single architectural floorplan and selected video walkthroughs. This allows simultaneous simulation of diverse robotic fleets operating in coordination:

• Humanoid robots navigating narrow corridors and performing dexterous assembly
• Autonomous Mobile Robots (AMRs) managing logistics and material transport
• Stationary robotic arms executing high-speed precision tasks

The platform also supports testing of multi-agent interactions on the same floor, ensuring collision-free synchronisation and optimised workflows before real-world deployment.

AuraML developed the platform on NVIDIA Omniverse and NVIDIA Isaac Sim, a suite of libraries and microservices for developing physical AI, digital twins and robotics. Powered by NVIDIA Cosmos world foundation models and accelerated computing, AuraML creates real-time, photorealistic environments that enable seamless data transfer between digital and physical systems.

In collaboration with NVIDIA, AuraML is launching The Global Physical AI & Robotics Cohort—an intensive, eight-week programme for robotics engineers, AI researchers and developers. With limited seats, the cohort offers a mentorship-driven experience focused on building, simulating and deploying real-world physical AI systems.

The launch marks a significant milestone in the evolution of Physical AI. By providing an end-to-end agentic workflow, AuraML enables robotics companies to move from concept to deployment with greater speed and efficiency.

“Our integrated workflows provide the critical infrastructure needed to make Physical AI more robust,” said Ayush Sharma, Co-founder of AuraML. “By launching the first multimodal world simulation model from India, powered by NVIDIA’s simulation technology, we are providing a global toolset to accelerate the development of advanced robotics foundation models.”

“India’s AI startup ecosystem is primed for acceleration, driven by exceptional technical talent and global ambition,” said Tobias Halloran, Director of EMEAI Startups and Venture Capital at NVIDIA. “NVIDIA is supporting this momentum by providing access to accelerated computing, scalable AI infrastructure, and programmes such as NVIDIA Inception and the Inception VC Alliance, helping startups scale and build for global markets.”

The announcement aligns with the momentum around the India–AI Impact Summit 2026, hosted by the Government of India under the IndiaAI Mission in New Delhi.