Kinara launches new edge AI processor to tackle generative AI demands

Kinara’s Ara-2 combines real-time responsiveness with high throughput, merging its latency optimized design with on-chip memories and high off-chip bandwidth to execute large models with low latency, according to the company.

“With Ara-2 added to our family of processors, we can better provide customers with performance and cost options to meet their requirements. For example, Ara-1 is the right solution for smart cameras as well as edge AI appliances with 2-8 video streams, whereas Ara-2 is strongly suited for handling 16-32+ video streams fed into edge servers, as well as laptops, and even high-end cameras,” says Ravi Annavajjhala, Kinara’s CEO.

“The Ara-2 enables better object detection, recognition, and tracking by using its advanced compute engines to process higher resolution images more quickly and with significantly higher accuracy. And as an example of its capabilities for processing generative AI models, Ara-2 can hit 10 seconds per image for Stable Diffusion and tens of tokens/sec for LLaMA-7B.”

Kinara’s Ara-2 aims to simplify the transition to the edge with its support for the 10’s of billions of parameters used by generative AI models.

Recently, Ampere welcomed Kinara into the AI Platform Alliance in a bid to reduce system complexity and promote collaboration with AI solutions. Ampere’s Chief Evangelist Sean Varley, says: “The performance and feature set of Kinara’s Ara-2 is a step in the right direction to help us bring better AI alternatives to the industry than the GPU-based status quo.”

The Ara-2 also offers encrypted memory access, and a host interface to enable enterprise AI deployments. Ara-2 is available as a stand-alone device, a USB module, an M.2 module, and a PCIe card featuring multiple Ara-2’s. Kinara will show a live demo with Ara-2 at CES.

Article Topics

edge AI  |  generative AI  |  Kinara  |  LLMs  |  processor