You are here:
4-/8-bit mixed-precision NPU IP
OPENEDGES, the world's only total memory system, and AI platform IP solution company, releases the first commercial mixed-precision (4-/8-bit) computation NPU IP, ENLIGHT. When ENLIGHT is used with other OPENEDGES' IP solutions, it reaches maximum efficiencies in power consumption, area, and DRAM optimization.
ENLIGHT, a high-performance neural network processor IP, features a highly optimized network model compiler that moves DRAM traffic from intermediate activation data by grouped layer partition and scheduling. Plus, it supports load balancing partition for multi-core NPU. With the industry's first adoption of 4-/8-bit mixed-quantization, it is easy to customize ENLIGHT at different core sizes and performance for the target market applications and achieve significant efficiencies in size, power, performance, and DRAM bandwidth.
A production-proven IP, ENLIGHT, has been licensed in a wide range of applications, including IP cameras, IoT, ADAS, and more.
ENLIGHT, a high-performance neural network processor IP, features a highly optimized network model compiler that moves DRAM traffic from intermediate activation data by grouped layer partition and scheduling. Plus, it supports load balancing partition for multi-core NPU. With the industry's first adoption of 4-/8-bit mixed-quantization, it is easy to customize ENLIGHT at different core sizes and performance for the target market applications and achieve significant efficiencies in size, power, performance, and DRAM bandwidth.
A production-proven IP, ENLIGHT, has been licensed in a wide range of applications, including IP cameras, IoT, ADAS, and more.
查看 4-/8-bit mixed-precision NPU IP 详细介绍：
- 查看 4-/8-bit mixed-precision NPU IP 完整数据手册
- 联系 4-/8-bit mixed-precision NPU IP 供应商
Block Diagram of the 4-/8-bit mixed-precision NPU IP
NPU IP
- Edge AI/ML accelerator (NPU)
- Neural network processor designed for edge devices
- Edge AI Accelerator NNE 1.0
- Neural Network Processor for Intelligent Vision, Voice, Natural Language Processing
- AI accelerator - 4.5K, 9K, or 18K INT8 MACs, 16 to 32TOPS
- AI accelerator - 36K or 54K INT8 MACs, 32 to 128TOPS