SiMa.ai Launches Modalix: A Game-Changing Chip for On-Device Physical AI

SiMa.ai Unveils Modalix: A Breakthrough in Physical AI Chip Technology

SiMa.ai, a US-based AI chip company, has announced the launch of its next-generation platform, Modalix, designed to accelerate the growth of Physical AI applications. This innovative Machine Learning System on a Chip (MLSoC) represents a significant leap forward in on-device AI processing capabilities, particularly for sectors such as robotics, automotive, and healthcare 1

1

2

2

.

Modalix: Powering the Era of Physical AI

The Modalix MLSoC is a second-generation chip that delivers high performance without compromising power efficiency. Operating at less than 10 watts, it enables real-time decision-making, natural language interaction, and complex AI model execution directly on edge devices 1

1

. This breakthrough allows for the running of various AI systems, including large language models (LLMs) like Meta's Llama, convolutional neural networks, transformer models, and vision language models (VLMs) 2

2

.

Krishna Rangasayee, founder and CEO of SiMa.ai, emphasized the significance of this launch, stating, "The era of Physical AI is here. With Modalix now in production, we're accelerating its global adoption" 1

1

.

Technical Specifications and Availability

The Modalix System-on-Module (SoM) is designed to match the pin configuration and form factor of leading graphics processing units, making it an ideal "brain" for edge devices running AI models 2

2

. SiMa.ai has introduced two variants of the Modalix SoM:

1. 8GB version: Starting at $349
2. 32GB version: Starting at $599

Additionally, a development kit priced at $1,499 is available for enterprise deployments of 1,000 units or more 2

2

.

LLiMa: Unifying AI Model Integration

Alongside Modalix, SiMa.ai has introduced LLiMa, a unified on-device framework for running LLMs, VLMs, and other AI models. This framework enables developers to quickly import both open-source and custom AI models into Modalix-ready binaries, while also supporting numerous libraries for physical AI applications 2

2

.

Industry Collaboration and Impact

SiMa.ai partnered with Synopsys Inc., a leader in electronic design and automation, to develop and innovate the new chip. This collaboration helped accelerate the design process and achieve bug-free silicon in record time 2

2

.

Ravi Subramanian, chief product management officer of Synopsys, highlighted the importance of this development, saying, "The development of Physical AI applications requires validated, purpose-built silicon and software that is only possible using the most advanced design solutions" 2

2

.

Applications and Future Prospects

The Modalix platform is poised to revolutionize various industries by enabling advanced AI capabilities in edge devices. Some potential applications include:

1. Automated visual inspection in manufacturing
2. Natural language command execution in robotics
3. Real-time decision-making in autonomous vehicles
4. Enhanced patient monitoring in healthcare

As the demand for edge AI continues to grow, SiMa.ai's Modalix platform is well-positioned to address the diverse needs of the industry, offering a versatile solution for integrating powerful AI capabilities into existing systems 2

2

.