Introduction:
Sima.ai, developed by Edge AI, announces that it has raised $80M in a Series B funding round led by Fidelity Management & Research Company. The startup series that the funds will be used to commercialize its first-generation system-on-a-chip product. It will also start the development of its second-generation product architecture. This supports the company’s go-to-market customer success and hiring initiatives globally.
Edge computing is forecasted to be a $6.72 billion market by 2022. Markets and Markets reported this forecast. The growth will coincide with the deep learning chipset market. Some analysts predict that it will reach $66.3 billion by 2025. There’s a reason for the rosy projections. Edge computing is expected to make roughly three-quarters of the total global AI chipset business in the coming six years.
Sima.ai unveils machine-learning-system-on-chip platform:
As it emerged from stealth in late 2019, Sima.ai unveils what it calls its machine-learning-system-on-chip platform. It is an AI accelerator chipset with low power requirements and supports fast inferencing. The company says that the hardware’s performance ranges from 50 TOPS to 200 TOPS at 5 Watts to 20 watts. It delivers what the company claims are an industry-first of 10 TOPS per watt.
Statement from Sima.ai VP of business development and system application, Sima.ai:
“[Our chip] combines traditional compute IP from Arm with our own machine learning accelerator and dedicated vision accelerator … By combining multiple machine learning accelerator mosaics via a proprietary interconnect, we can scale from 50 TOPs at 5 Watts up to [a theoretical] 400 TOPs at 40 Watts,” Kavitha Prasad, VP of business development and system applications at Sima.ai, explained in a blog post last year. “While it’s capable of a wide range of ML workloads such as natural language processing, SiMa.ai’s [chip] is initially optimized for computer vision applications.”
Sima.ai plans to work with customers in robotics, smart cities, autonomous vehicles, medical imaging, and government. It also claims to have completed several early customer engagements. Recently, it announced the opening of a design center in Bengaluru, India. This center will support engineering and operations.
It will launch job opportunities for board development, operations, infrastructure, and system application roles. “The embedded edge is a multi-trillion dollar market and still using decades-old technology. Sima.ai is poised to disrupt this massive market with our differentiated machine learning technology and approach,” founder and CEO Krishna Rangasayee said in a press release.
Competition:
It should be noted that Sima.ai has a lot of things in the way of competition. Startups AIstorm, Hailo, Esperanto technologies, Quadric, Graphcore, Xnor, and Flex Logix are developing chips customized for AI workloads.
Mobileye, the Tel Aviv company Intel, was acquired for $15.3 billion in March 2017. This company offers a computer vision processing solution for AVs in its EyeQ product line. By 2 last July unveiled Kunlun, which is a chip for edge computing on devices and in the cloud via data centers.
But Sima.ai appears well capitalized with $120 million in funding to date. It also closed a $30 million financing round in May 2020 led by Dell Technologies Capital. The company plans to tape out its chipset early this year with the goal to deliver engineering samples. It would also include customer samples by the end of 2021.
