Seeed Studio BeiQi RK3399Pro AIoT 96Boards Compute SoM is a powerful core board for Artificial Intelligence (AI) that can form a complete development or evaluation board. This board uses the RK3399Pro as the main control chip. The BeiQi RK3399Pro AIoT board features a six-core 64-bit processor, two CPU clusters, 8-bit/16-bit operation support, powerful multimedia processing performance, and strong hardware codec capability. The CPU clusters include a little cluster with quad-core Cortex-A53 optimized for low power and a big cluster with dual-core Cortex-A72 optimized for high-performance. This board also offers rich expansion interfaces, AI application development interface, a high-speed onboard connector for stability and reliability, and multiple OS support. The BeiQi RK3399Pro AIoT board also features dual-core ARM Cortex-A72 and quad-core ARM Cortex-A53 MPcore processors which are high-performance, low-power, and cache application processors.
Features
Six-core 64-bit processor for superior general-purpose computing powerDual-core ARM Cortex-A72 MPcore processor and quad-core ARM Cortex-A53 MPcore processor are high-performance, low-power, and cache application processorsTwo CPU clusters:Big cluster with dual-core Cortex-A72 optimized for high-performanceLittle cluster with quad-core Cortex-A53 optimized for low powerFull implementation of the ARM architecture v8-A instruction set, ARM Neon Advanced Single Instruction, Multiple Data (SIMD) support for accelerating media and signal processingSupports 8-bit/16-bit operation, AI computing power up to 3.0 TOPs (INT8 Inference)Full load calculation is strong and light load operation power consumption is lowCompatible with Caffe/Mxnet/TensorFlow model, it can support multiple frameworks, support mainstream layer types, and add custom layer easilyProvide easy-to-use development tools, PC can complete model conversion, performance prediction, accuracy verificationProvide AI application development interface:Supports Android NN API, RKNN cross-platform API, and Linux support TensorFlow developmentPowerful multimedia processing performanceIntegrated quad-core ARM Mali-T860 MP4 GPU
Specifications
Rockchip RK3399Pro SoC3GB LPDRR3（CPU 2GB + NPU 1GB）RAM16GB eMMC
Applications
Industrial automationUAVImage detectionFace recognitionEdge computing gatewayCluster server
