Is this your business? Claim it to manage your IP and profile
SAKURA-II AI Accelerator represents EdgeCortix's latest advancement in edge AI processing, offering unparalleled energy efficiency and extensive capabilities for generative AI tasks. This accelerator is designed to manage demanding AI models, including Llama 2, Stable Diffusion, DETR, and ViT, within a slim power envelope of about 8W. With capabilities extending to multi-billion parameter models, SAKURA-II meets a wide range of edge applications in vision, language, and audio. The SAKURA-II's architecture maximizes AI compute efficiency, delivering more than twice the utilization of competitive solutions. It boasts remarkable DRAM bandwidth, essential for large language and vision models, while maintaining low power consumption. The hardware supports real-time Batch=1 processing, demonstrating its edge in performance even in constrained environments, making it a choice solution for diverse industrial AI applications. With 60 TOPS (INT8) and 30 TFLOPS (BF16) in performance metrics, this accelerator is built to exceed expectations in demanding conditions. It features robust memory configurations supporting up to 32GB of DRAM, ideal for processing intricate AI workloads. By leveraging sparse computing techniques, SAKURA-II optimizes its memory and bandwidth usage effectively, ensuring reliable performance across all deployed applications.
The Dynamic Neural Accelerator II (DNA-II) from EdgeCortix is an advanced neural network IP core tailored for high efficiency and parallelism at the edge. Incorporating a run-time reconfigurable interconnect system between compute units, DNA-II effectively manages both convolutional and transformer workloads. This architecture ensures scalable performance beginning with 1K MACs, suitable for a wide range of SoC implementations. EdgeCortix's patented architecture significantly optimizes data paths between DNA engines, enhancing parallelism while reducing on-chip memory usage. As a core component of the SAKURA-II platform, DNA-II supports state-of-the-art generative AI models with industry-leading energy efficiency. DNA-II's design acknowledges the typical inefficiencies in IP cores, improving compute utilization and power consumption metrics substantially. By adopting innovative reconfigurable datapath technologies, EdgeCortix sets a new benchmark for low-power, high-performance edge AI applications.
The MERA Compiler and Software Framework by EdgeCortix is designed to streamline AI model deployment across various platforms, simplifying the transition of pre-trained models to functional applications. By offering a robust set of tools for model calibration and quantization, MERA ensures that models run optimally on the Dynamic Neural Accelerator (DNA) architecture. MERA's integration capabilities extend across leading processors such as AMD, Intel, Arm, and RISC-V, providing flexibility and ease of use. It supports a range of machine learning frameworks like TensorFlow Lite and ONNX, making it easier for developers to adapt to this platform. The framework is especially suited for edge AI tasks where quick adaptation and efficiency are necessary. Apart from model portability, MERA stands out with its ease of integration, supporting popular neural network models, including those from Hugging Face. Its open-source front end ensures community involvement and adaptability, encouraging industry-wide collaboration and innovation in AI deployment strategies.
EdgeCortix secures series B funding, nearing $100M. Discover their strategic expansion in AI processing for diverse industries. Insights from key investors. Read more
Explore the breakthrough collaboration between EdgeCortix and Renesas with their RUHMI Framework, simplifying AI deployment on MCUs and MPUs. Read more
EdgeCortix to advance the NovaEdge AI chiplet with NEDO funding for energy-efficient edge solutions in AI, set to revolutionize multiple tech sectors. Read more
Join the world's most advanced semiconductor IP marketplace!
It's free, and you'll get all the tools you need to discover IP, meet vendors and manage your IP workflow!
No credit card or payment details required.
Join the world's most advanced AI-powered semiconductor IP marketplace!
It's free, and you'll get all the tools you need to advertise and discover semiconductor IP, keep up-to-date with the latest semiconductor news and more!
Plus we'll send you our free weekly report on the semiconductor industry and the latest IP launches!
To evaluate IP you need to be logged into a buyer profile. Select a profile below, or create a new buyer profile for your company.
