Is this your business? Claim it to manage your IP and profile
The Dynamic Neural Accelerator II Architecture (DNA-II) by EdgeCortix is a sophisticated neural network IP core structured for extensive parallelism and efficiency enhancement. Distinguished by its run-time reconfigurable interconnects between computing elements, DNA-II supports a broad spectrum of AI models, including both convolutional and transformer networks, making it suitable for diverse edge AI applications. With its scalable performance starting from 1K MACs, the DNA-II architecture integrates easily with many SoC and FPGA applications. This architecture provides a foundation for the SAKURA-II AI Accelerator, supporting up to 240 TOPS in processing capacity. The unique aspect of DNA-II is its utilization of advanced data path configurations to optimize processing parallelism and resource allocation, thereby minimizing on-chip memory bandwidth limitations. The DNA-II is particularly noted for its superior computational capabilities, ensuring that AI models operate with maximum efficiency and speed. Leveraging its patented run-time reconfigurable data paths, it significantly increases hardware performance metrics and energy efficiency. This capability not only enhances the compute power available for complex inference tasks but also reduces the power footprint, which is critical for edge-based deployments.
The SAKURA-II AI Accelerator represents the pinnacle of EdgeCortix's innovation, offering cutting-edge efficiency in AI inferencing tasks particularly suited for generative AI applications. This high-performance accelerator is powered by the low-latency Dynamic Neural Architecture (DNA), handling multi-billion parameters with remarkable adequacy. Its adaptability allows it to transform inputs across various modalities such as vision, language, and audio. Particularly distinguished for its compactness and minimal power consumption, the SAKURA-II is engineered to deliver excellent results even in constrained physical environments. It supports vast model complexities with impressive DRAM bandwidth and is equipped for low-latency operations, ensuring fast real-time Batch=1 processing. Notably, its hardware capacity efficiently approximates activation functions, extending substantial support for structures like Llama 2, Stable Diffusion, and ViT. Critically, the SAKURA-II is designed for flexibility, offering robust processing of vision and generative AI workloads in an 8W power envelope. It features a highly-operative memory infrastructure that enhances its performance with sophisticated DRAM bandwidths and ample memory capacity for handling complex data streams and facilitating superior AI compute utility. Its intelligent design aligns with the current move towards sophisticated, low-power AI applications.
The MERA Compiler and Framework is a groundbreaking software platform designed by EdgeCortix to streamline the deployment of sophisticated neural network models across heterogeneous systems. Tailored to handle complex generative AI tasks, MERA provides a framework-agnostic base for model porting and system integration, facilitating the compilation and inference of deep neural networks. MERA stands out for its comprehensive support for major ML frameworks including PyTorch, TensorFlow Lite, and ONNX, leveraging native API compatibility for seamless workflow integration. Developers benefit from a rich suite of tools and code generators that optimize neural network models for edge deployment, including functionalities for post-training calibration and quantization. Designed to foster innovation, the platform offers diverse benefits such as the integration of Apache TVM and MLIR for efficient code generation. The framework enhances productivity by allowing users to utilize pre-trained models sourced from EdgeCortix's Model Library or Hugging Face. MERA, working in conjunction with the Dynamic Neural Accelerator IP, supports edge AI deployments with reduced time-to-market and enables real-time, low-latency application execution.
Join the world's most advanced semiconductor IP marketplace!
It's free, and you'll get all the tools you need to evaluate IP, download trial versions and datasheets, and manage your evaluation workflow!
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.
