VAST BOX

Leetop vast Xavier is an embedded artificial intelligence computer, which can give up to 32tops computing power to many terminal devices. Leetop vastxavier provides fast passive heat dissipation design, which can meet industrial standards such as earthquake resistance and anti-static. At the same time

The option marked with * is required. After submission, our company will arrange a business email to send you the corresponding product PDF document

Product features

Product function

Leetop Vast-XX Box

Nvidia Jetson Xavier / Orin

Take supercomputer performance to the edge.


Small Size. Small Price. Big AI Discoveries. 


Leetop VAST BOX, based on discover the power of AI and robotics with NVIDIA® Jetson . It’s small, powerful, and priced for everyone . This means educators, students, and other enthusiasts can now easily create projects with fast and efficient AI using the entire GPU-accelerated NVIDIA software stack.

The Leetop VAST BOX benefits from new cloud-native support across the entire Jetson platform line-up, making it easier to build, manage, and deploy AI at the edge.

Pre-trained AI models from NVIDIA NGC together with the NVIDIA Transfer Learning Toolkit, provides a faster path to trained and optimized AI networks. Containerized deployment to Jetson devices also allows flexible and seamless updates.

NVIDIA JetPack™ SDK enables multi-modal AI application development for Leetop VAST BOX with accelerated libraries supporting all major AI frameworks, as well as computer vision, computer graphics, multimedia, and more. Together with the latest NVIDIA tools for application development and optimization, JetPack ensures fast time to market and reduced development costs.

Designed for ease of development and speed of deployment, Leetop is the most flexible and scalable platform to get to market and continuously update AI software over the lifetime of a product.


VAST BOX(图1)Interface



Interface

Specification


Network

2 x Gigabit Ethernet Connector (10/100/1000)


Video Output

2 x HDMI 2.0 (TYPE A)


USB

6 x USB 3.0 (TYPE A) + 1 x USB-OTG (Type C)


Audio

1 x Audio Jack (3.5mm)


M.2 KEY M

1x M.2 KEY M (NVMe SSD)


TF Slot

1 x TF Card Slot


SIM Slot

1 x SIM Card Slot


Function Keys

1 x Reset Button + 1 x Recovery Button + 1 x Power Button


Serial port (Optional)

RS232 x 1 (2)   / RS485 x 1 (2)   / 1X SPI Bus(+3.3V Level) / 1x CAN /2x I2C Link

(+3.3V I/O) 1x UART(+3.3V Level) / 1X SPI Bus(+3.3V Level) / 2x GPIO(+3.3V

Level)




KEY FEATURES


Processor

NVIDIA Jetson AGX Xavier

NVIDIA Jetson AGX Orin

AI Performance

32 TOPs

Up to 200 Sparse TOPS (INT8)

 

GPU

 

512-core Volta GPU with 64 Tensor Cores 11 TFLOPS (FP16) 22 TOPS (INT8)

2 GPC/ 8 TPC/MIG/2048 NVIDIA® CUDA® cores / 64

Tensor cores Ray-Tracing cores/ End-to-end lossless compression / Tiled Caching / OpenGL® 4.6+ / OpenGL ES 3.2/ Vulkan™ 1.2+◊/

GPU Max Freq

2200 MHz

1 GHz

CPU

8-core Carmel ARM v8.2 64-bit CPU, 8MB L2 + 4MB L3

Arm Cortex-A78AE

CPU Max Freq

4-core @ 2200MHz 6/8-core @ 1400MHz

2 GHz

Memory

32GB 256-Bit LPDDR4x |136.5GB/s

32 GB 256-bit LPDDR5 DRAM

Storage

32GB eMMC 5.1

64 GB eMMC 5.1 64 MB NOR Boot Flash 8 MB NOR

Secure Key Flash

Power

10W |15W |30W

5V and 20V

PCIe

1 x8 + 1 x4 + 1 x2 + 2 x1 (PCIe Gen4, Root Port & Endpoint)

PCI Express 4.0 x1, x2, x4 and x8 Up to 3x Endpoint supported

 

CSI Camera

Up to 6 cameras (36 via virtual channels) 16 lanes MIPI CSI-2 |8 lanes SLVS-EC D-PHY 1.2 (up to 40 Gbps) C-PHY 1.1 (up to 62 Gbps)

 

16×: CSI Lanes D-PHY v2.1

Video Encode

2x1000MP/sec 4x 4K @ 60 (HEVC) 8x 4K @ 30 (HEVC) 16x 1080p @ 60 (HEVC) 32x 1080p @ 30 (HEVC)

2×: 4K60 (H.265) /1x: 8K30 (H.265)

 

Video Decode

2x1500MP/sec 2x 8K @ 30 (HEVC) 6x 4K @ 60 (HEVC) 12x 4K @ 30 (HEVC) 26x 1080p @ 60 (HEVC) 52x 1080p @ 30 (HEVC) 30x 1080p @ 30 (H.264)

 

2×: 4K60 (H.265) /1x: 8K30 (H.265)

Display

3 multi-mode DP 1.4/eDP 1.4/HDMI 2.0 No DSI support

1×: shared HDMI™ 2.0, eDP1.4, VESA® DisplayPort™ (DP) HBR3

DL Accelerator

(2x) NVDLA Engines* |5 TFLOPS (FP16), 10 TOPS (INT8)

Deep Learning Accelerator (DLA) / Up to 90 INT8 TOPS (Deep Learning Inference) / 2MB dedicated SRAM

 

Vision Accelerator

 

7-way VLIW Vision Processor*

Programmable Vision Accelerator (PVA) / Up to 512 INT16 GMACS or 2048 INT8 GMACS / 2MB dedicated SRAM

Networking

10/100/1000 BASE-T Ethernet

RGMII

Mechanical

100 mm x87 mm 699-pin connector Integrated Thermal Transfer Plate

100.0 mm x 87.0 mm x 16.0 mm 699 pin B2B Connector Integrated Thermal Transfer Plate (TTP) with Heatpipe


Install Dimension


VAST BOX(图2)



Online consultation