SiMa.ai Launches Chip to Run Reasoning-Based LLMs On-Device in Under 10 Watt

The new Modalix SoM and DevKit are now available, with pricing for the 8GB SoM starting at $349 and the 32GB version at $599.

SiMa.ai, the US-based AI chip company, has launched its next-generation platform to accelerate the growth of Physical AI applications. The company introduced the Modalix Machine Learning System on a Chip (MLSoC) along with a new System-on-Module (SoM) and development kits, aiming to meet the growing demand for Physical AI in sectors like robotics, automotive, and healthcare.

Modalix is a second-generation MLSoC designed to deliver high performance without compromising power efficiency, operating at less than 10 watts. “The era of Physical AI is here. With Modalix now in production, we’re accelerating its global adoption,” said Krishna Rangasayee, founder and CEO of SiMa.ai.

Its flexible, Arm-based architecture enables real-time decision-making, natural language interaction, and seamless support for large language models (LLMs), transformers, convolutional neural networks (CNNs), and generative AI (GenAI) workloads.

Seamless Integration 

SiMa.ai also unveiled its new Modalix SoM, which it claims is pin-compatible with leading GPU SoMs. This feature enables easy integration into existing systems, making it an attractive option for developers. 

The platform includes MIPI, memory, and other essential I/O components for rapid scaling of Physical AI systems. The company’s LLiMa software framework, which supports LLM deployment, further simplifies integrating GenAI capabilities into Physical AI applications.

“Modalix showcases the scale of innovation possible with Arm’s flexible, power-efficient compute platform,” said Ami Badani, chief marketing officer of Arm. “SiMa.ai is enabling smarter and more sustainable systems across industries.”

SiMa.ai collaborated with Synopsys to speed up development, using the company’s AI-powered design tools. TSMC’s advanced N6 process ensured Modalix met power, thermal, and reliability standards.

Sajiv Dalal, president of TSMC North America, said, “This collaboration underscores our commitment to driving energy-efficient chip innovations that are redefining the future of AI.”

Global Launch and Commercial Availability

The new Modalix SoM and DevKit are now available, with pricing for the 8GB SoM starting at $349 and the 32GB version at $599. The DevKit is priced at $1,499. SiMa.ai aims to accelerate the global adoption of Physical AI, with strong demand for its products.

In a recent exclusive interview with AIM, Rangasayee also praised the Indian market, saying, “We believe India could be the next market maker for the planet. And not only for local consumption, but for global consumption.”The company recently raised $85 million in an oversubscribed funding round, taking its total capital raised to $355 million. It had hinted that the funds would be used to expand globally and scale its physical AI platform.

📣 Want to advertise in AIM? Book here

Picture of Sanjana Gupta
Sanjana Gupta
An information designer by training, Sanjana likes to delve into deep tech and enjoys learning about quantum, space, robotics and chips that build up our world. Outside of work, she likes to spend her time with books, especially those that explore the absurd.
Related Posts
AIM Print and TV
Don’t Miss the Next Big Shift in AI.
Get one year subscription for ₹5999
Download the easiest way to
stay informed