The demand for energy-efficient AI acceleration hardware with cheap capital costs is growing as AI advances in every discipline. Adding ML to existing products is difficult as ML solutions are not purpose-built for the embedded edge. The need for specialized Chipsets to boost AI workloads on-premises is increasing day by day.
SiMa.ai is working on a system-on-chip platform for artificial intelligence (AI) applications, specifically computer vision. They are speeding up the adoption of high-performance machine learning inference in embedded edge applications while using significantly less power.
The tech firm is making it easier to apply machine learning to various devices by offering its software-centric purpose-built MLSoC platform with push-button performance for easy ML deployment and scaling at the embedded edge. The MLSoC is ideal as a standalone edge-based system controller or as a machine learning offload accelerator for processors, ASICs, and other devices due to its low running power and high ML processing capability. The company claims that its platform is ten times better than the alternatives available in the market.
What sets SiMa.ai apart from its competitors is that most companies start with a silicon-centric emphasis and work on software later, but SiMa.ai discovered that power efficiency and software ease-of-use are even more critical. Hence, it focuses on these. Their accelerator chipset combines Arm’s “conventional computing IP” with a unique machine learning accelerator and a dedicated vision accelerator, both of which are connected through a proprietary connection. They intend to collaborate with firms working on self-driving vehicles, robotics, medical equipment, drones, and other technologies.
The company has raised $30 million in a Series B funding round. This funding brings the company’s total capital raised to $150 million. SiMa.ai will use these funds to expand the engineering and business teams globally and continue investing in hardware and software innovation.
Pitch your startup story at [email protected]
Please don't forget to join our ML Subreddit