Sima.ai unveils two PCIe-based production boards to ease edge AI development
Sima.ai, a provider of machine learning system-on-chip platforms, has launched PCI express half-height half-length (PCIe HHHL) and Dual M.2 production boards built with the MLSoC platform. The company states that the onboard system-on-chip can enhance performance by a factor of ten and can continually innovate via new algorithms and neural networks after deployment.
The PCIe HHHL and Dual M.2 are incredibly versatile production boards, enabling customers to swiftly enter the production phase without waiting for lengthy internal board development cycles. The board designs also allow customers to develop custom form factors.
The Sima.ai machine learning system-on-chip offers a computational performance of up to 50 tera operations per second for neural networks. The hardware includes an application processing unit with four Arm Cortex-A65 dual processor cores running at a frequency of 1.15GHz. It also features four Synopsys ARC EV74 video processors for computer vision processing that can handle up to 600 16-bit GOPS.
“We’re excited to bring these new form factor boards, programmed with our Palette software, to market for our customers because they address a growing need for a combined complete software and hardware solution within the developer community,” said Krishna Rangasayee, the CEO and founder of SiMa.ai.
These hardware platforms are now available for commercial use.
Sima.ai debuts Palette software platform to simplify ML application development
Sima.ai has launched its Palette software development platform, offering a seamless user experience for the development of machine learning applications at the embedded edge, the company says. According to Sima.ai, this platform solves scalability issues and supports a range of machine-learning algorithms.
The machine learning compiler within the Palette software allows for the integration of models from any neural network framework. The software development platform enables users to design a machine learning workflow, build integrated applications, deploy embedded machine learning and debug various applications in real-time.
“Palette is ML software done right, delivering on the promise to developers, simultaneously delivering the flexibility, performance, and ease-of-use that we have committed to in our any network, 10x performance, pushbutton strategy for accelerating edge ML deployment,” said Rangasayee.
Late last year, Sima.ai secured $67 million in a Series B1 funding round to help advance its product line. The company intended to use the funds to develop its MLSoC platform further and create a second-generation product design.
Article Topics
AI/ML | edge AI | model development | neural networks | PCI | Sima.ai
Comments