Nano Module

The Nano moduledelivers up to 472 GFLOPS of accelerated computing from CUDA cores in an unprecedented size, power and cost. nano natively supports all types of AI models and popular AI frameworks e.g. TensorFlow, PyTorch and MxNet. This makes nano the ideal platform for developing mass market AI products such as AIIoT gateways, smart network video recorders and cameras, consumer robots, and optical inspection systems.

The system-on-module is powered by the Maxwell GPU with 4GB of memory which enables real-time processing of high resolution inputs. It has a combination of performance, power advantage and rich set of IOs, from high speed CSI and PCIe to low speed I2Cs and GPIOs. It can also support multiple diverse sets of sensors to enable a variety of applications with incredible power efficiency, consuming as little as 5W.

Nano is supported byJetPack, which includes a board support package (BSP), Linux OS, CUDA , cuDNN, and TensorRT software libraries for deep learning, computer vision, GPU computing, multimedia processing, and much more. The comprehensive software stackmakes AI deployment on autonomous machines fast, reduces complexity and time to market.

• Small size
• High computational performance
• low power

• GPU:Maxwell architecture with 128 CUDA cores0.5 TFLOPs (FP16)
• CPU: Quad-core ARM Cortex -A57 MPCore processor
• Memory: 4 GB 64-bit LPDDR4 1600MHz - 25.6 GB/s
• Storage: 16 GB eMMC 5.1 Flash
• Video Encode: 250 MP/sec | 1x 4K @ 30 (HEVC) | 2x 1080p @ 60 (HEVC) | 4x 1080p @ 30 (HEVC)
• Video Decode: 500 MP/sec | 1x 4K @ 60 (HEVC) | 2x 4K @ 30 (HEVC) | 4x 1080p @ 60 (HEVC) | 8x 1080p @ 30 (HEVC)
• Camera: 12 lanes (3x4 or 4x2) MIPI CSI-2 D-PHY 1.1 (18 Gbps)
• Display: HDMI 2.0 or DP1.2 | eDP 1.4 | DSI (1 x2) 2 simultaneous
• UPHY: 1 x1/2/4 PCIE, 1x USB 3.0, 3x USB 2.0
• I/O: 3x UART, 2x SPI, 2x I2S, 4x I2C, GPIOs
• Size: 69.6 mm x 45 mm
• Mechanical: 260-pin edge connector