tsunAImi accelerator card

tsunAImi accelerator card

Software for AI Acceleration/Optimization

Information

Untether AI’s at-memory compute architecture breaks through the computational bottleneck and changes the paradigm for AI compute efficiency. Untether AI’s tsunAImi accelerator cards powered by 4 runAI200 devices provide record-breaking energy efficiency and compute density for inference acceleration. The tsunAImi accelerator cards provide 2 PetaOps of compute power, more than twice the amount of any available product. Natively a batch=1, tsunAImi cards are equally adept at running neural networks for vision, natural language processing/understanding, and recommendation engines.
UNTETHER AIUntether AI provides ultra-efficient, high-performance AI chips to enable new frontiers in AI applications. By combining the power efficiency of at-memory computation with the robustness of digital processing, Untether AI has developed a ground-breaking new chip architecture for neural net inference that eliminates the data movement bottleneck that costs energy and performance in traditional architectures. 

Log in