HomeProduct CenterNVIDIA GPU AI Computing

NVIDIA GPUAI Computing

From professional graphics to data centers, from edge computing to AI inference, NVIDIA's full product portfolio delivers powerful computing power for every industry.

No category data available
Shared24This product
NVIDIA Vera Rubin NVL72
AI

NVIDIA Vera Rubin NVL72

The NVIDIA Vera Rubin NVL72 integrates 72 NVIDIA Rubin GPUs, 36 NVIDIA Vera CPUs, NVIDIA ConnectX-8 SuperNICs, and NVIDIA BlueField-4 DPUs into a single rack-scale platform. Powered by sixth-generation NVIDIA NVLink and NVLink Switch, along with NVIDIA Quantum-X800 InfiniBand and NVIDIA Spectrum-X800 Ethernet, it delivers the compute power needed to fuel the AI revolution.

GPU Architecture
Vera Rubin
Number of GPUs
72
CPU
NVIDIA Vera CPU
In StockStock:99
NVIDIA GeForce RTX 5090
RTX Ada/Quadro

NVIDIA GeForce RTX 5090

NVIDIA GeForce RTX 5090 flagship GPU with Blackwell architecture delivers ultimate performance for gamers and creators.

architecture
Blackwell
cuda cores
21760
memory
3 2GB GDDR7
In Stock
NVIDIA Jetson AGX Orin 64GB Developer Kit
Jetson

NVIDIA Jetson AGX Orin 64GB Developer Kit

The NVIDIA Jetson AGX Orin 64GB Developer Kit is a flagship embedded computing platform designed for autonomous machines and edge AI applications. Powered by the Orin SoC, it delivers up to 275 TOPS of AI performance, features 64GB LPDDR5 memory, supports multi-camera input, and offers rich connectivity—ideal for robotics, intelligent video analytics, industrial automation, and more.

AI Compute
275 TOPS
memory
6 4GB LPDDR5
GPU Architecture
Ampere
¥19,999¥21,999
SKU:NVIDIA JETSON AGX ORIN 64G
In StockStock:99
NVIDIA Jetson AGX Orin 32GB Developer Kit
Jetson

NVIDIA Jetson AGX Orin 32GB Developer Kit

The NVIDIA Jetson AGX Orin 32GB Developer Kit delivers up to 200 TOPS of AI performance with 32GB LPDDR5 memory, making it a high-performance embedded platform for advanced edge AI applications. Ideal for autonomous navigation, intelligent robotics, and edge inference scenarios.

AI Compute
200 TOPS
memory
3 2GB LPDDR5
GPU Architecture
Ampere
¥14,999¥16,999
SKU:NVIDIA-JETSON-AGX-ORIN-32G
In StockStock:99
NVIDIA Groq 3 LPX
AI

NVIDIA Groq 3 LPX

The NVIDIA Groq 3 LPX is an inference accelerator designed for NVIDIA Vera Rubin, optimized for low-latency, large-context agent systems and large-scale content generation. Each rack supports 256 LPX units, delivering 128GB of SRAM, 40PB/s of memory bandwidth, and 640TB/s of rack-level scaling bandwidth.

GPU Architecture
Vera Rubin
LPX Quantity
256 per rack
SRAM
128GB
In StockStock:99
NVIDIA Jetson Orin NX 16GB Developer Kit
Jetson

NVIDIA Jetson Orin NX 16GB Developer Kit

The NVIDIA Jetson Orin NX 16GB Developer Kit delivers up to 100 TOPS of AI performance with 16GB LPDDR5 memory, offering powerful edge AI in a compact module form factor. Ideal for drones, mobile robots, and smart cameras.

AI Compute
100 TOPS
memory
16GB LPDDR5
GPU Architecture
Ampere
¥5,999¥6,999
SKU:NVIDIA-JETSON-ORIN-NX-16G
In StockStock:99
NVIDIA DGX Vera Rubin NVL72
AI

NVIDIA DGX Vera Rubin NVL72

NVIDIA DGX Vera Rubin NVL72 Integrated AI Supercomputing Platform Seamlessly Integrates Vera Rubin GPUs with Grace CPUs.

architecture
Vera Rubin + Grace
GPU count
72
cpu
Grace ARM
In Stock
NVIDIA GeForce RTX 5070 Ti
RTX Ada/Quadro

NVIDIA GeForce RTX 5070 Ti

NVIDIA GeForce RTX 5070 Ti, built on the Blackwell architecture with GDDR7 memory, delivers exceptional performance for gamers and creators. Supports DLSS 4 Frame Generation, full ray tracing, and Reflex low-latency technology.

GPU Architecture
Blackwell
VRAM
1 6GB GDDR7
CUDA Cores
8960
¥6,499¥6,999
SKU:NVIDIA RTX 5070 Ti-16G
In StockStock:99
NVIDIA HGX Rubin NVL8
AI

NVIDIA HGX Rubin NVL8

The NVIDIA HGX platform unites the power of NVIDIA GPUs, NVLink networking, and a fully optimized AI and high-performance computing (HPC) software stack to deliver peak application performance and fastest insights for every data center.

GPU Architecture
Rubin
Number of GPUs
8
Interconnect
NVLink
In StockStock:99
NVIDIA DGX Rubin NVL8
AI

NVIDIA DGX Rubin NVL8

The NVIDIA DGX Rubin NVL8 is a liquid-cooled AI system featuring eight NVIDIA Rubin GPUs and sixth-generation NVLink, designed to accelerate training, inference, and post-inference processing for diverse AI workloads.

GPU Architecture
Rubin
Number of GPUs
8
Interconnect
6th Gen NVLink
In StockStock:99
NVIDIA RTX PRO 6000 Blackwell Server Edition
AI

NVIDIA RTX PRO 6000 Blackwell Server Edition

Professional server GPU for enterprise AI workloads, based on Blackwell architecture and supporting Multi-Instance GPU (MIG) technology.

architecture
Blackwell
cuda cores
18432
memory
9 6GB GDDR7
In Stock
NVIDIA GeForce RTX 5070
RTX Ada/Quadro

NVIDIA GeForce RTX 5070

The NVIDIA GeForce RTX 5070, built on the Blackwell architecture and featuring GDDR7 memory, delivers an exceptional gaming and creation experience. With support for DLSS 4 Frame Generation technology, it significantly boosts in-game frame rates.

GPU Architecture
Blackwell
VRAM
1 2GB GDDR7
CUDA Cores
6144
¥4,599¥4,999
SKU:NVIDIA-RTX5070-12G
In StockStock:99
NVIDIA Jetson Orin NX 8GB Developer Kit
Jetson

NVIDIA Jetson Orin NX 8GB Developer Kit

The NVIDIA Jetson Orin NX 8GB Developer Kit delivers up to 70 TOPS of AI performance with 8GB of LPDDR5 memory, offering an exceptional value for edge AI computing. Ideal for smart retail, lightweight robotics, and edge inference applications.

AI Compute
70 TOPS
memory
8GB LPDDR5
GPU Architecture
Ampere
¥3,999¥4,499
SKU:NVIDIA-JETSON-ORIN-NX-8G
In StockStock:99
NVIDIA Vera CPU
AI

NVIDIA Vera CPU

NVIDIA Vera is built for reinforcement learning and agent AI, delivering compute power for code, tool, and data workflows. As the first CPU in a GPU-accelerated system, Vera seamlessly collaborates with NVIDIA GPUs to orchestrate data flow, manage memory, and coordinate system control—keeping your AI pipeline running at full speed.

architecture
ARM
Usage
Reinforcement Learning and Agent AI
Features
Seamlessly integrates with NVIDIA GPUs
In StockStock:99
NVIDIA GeForce RTX 5060 Ti
RTX Ada/Quadro

NVIDIA GeForce RTX 5060 Ti

The NVIDIA GeForce RTX 5060 Ti, built on the Blackwell architecture, delivers exceptional value for mainstream gamers. It supports advanced features like DLSS 4 and full ray tracing, performing exceptionally well at 1080p and 1440p resolutions.

GPU Architecture
Blackwell
VRAM
1 6GB GDDR7
CUDA Cores
4608
¥3,299¥3,499
SKU:NVIDIA RTX 5060 Ti-16G
In StockStock:99
NVIDIA GeForce RTX 5080
RTX Ada/Quadro

NVIDIA GeForce RTX 5080

NVIDIA GeForce RTX 5080 High-Performance GPU with Blackwell Architecture, built for gaming and content creation.

architecture
Blackwell
cuda cores
10752
memory
1 6GB GDDR7
In Stock
NVIDIA Jetson Orin Nano 8GB Developer Kit
Jetson

NVIDIA Jetson Orin Nano 8GB Developer Kit

The NVIDIA Jetson Orin Nano 8GB Developer Kit delivers up to 40 TOPS of AI performance with 8GB of LPDDR5 memory, making it the ideal choice for entry-level edge AI development. It supports the full NVIDIA AI software stack and is designed for smart cameras, IoT devices, and lightweight AI applications.

AI Compute
40 TOPS
memory
8GB LPDDR5
GPU Architecture
Ampere
¥2,499¥2,999
SKU:NVIDIA-JETSON-ORIN-NANO-8G
In StockStock:99
NVIDIA Jetson Orin Nano 4GB Developer Kit
Jetson

NVIDIA Jetson Orin Nano 4GB Developer Kit

The NVIDIA Jetson Orin Nano 4GB Developer Kit delivers up to 20 TOPS of AI performance with 4GB of LPDDR5 memory, offering the most cost-effective entry point into the Jetson Orin series. Ideal for education, prototyping, and lightweight edge AI applications.

AI Compute
20 TOPS
memory
4GB LPDDR5
GPU Architecture
Ampere
¥1,499¥1,799
SKU:NVIDIA-JETSON-ORIN-NANO-4G
In StockStock:99
NVIDIA GB300 NVL72
AI

NVIDIA GB300 NVL72

NVIDIA GB300 NVL72 integrates 36 NVIDIA Grace CPUs and 72 NVIDIA Blackwell Ultra GPUs into a single rack-level design with NVIDIA NVLink for high-speed interconnectivity in AI inference, training, and data processing. Built specifically for AI inference and large language model (LLM) inference, GB300 NVL72 delivers breakthrough performance.

GPU Architecture
Blackwell Ultra
Number of GPUs
72
CPU
36 Grace CPUs
In StockStock:99
NVIDIA GeForce RTX 5060
RTX Ada/Quadro

NVIDIA GeForce RTX 5060

The NVIDIA GeForce RTX 5060, built on the Blackwell architecture with GDDR7 memory, is ideal for entry-level gamers. It supports DLSS 4 Frame Generation and full-ray tracing for a smooth 1080p gaming experience.

GPU Architecture
Blackwell
VRAM
8GB GDDR7
CUDA Cores
3840
¥2,499¥2,699
SKU:NVIDIA-RTX5060-8G
In StockStock:99
NVIDIA GeForce RTX 4090
RTX Ada/Quadro

NVIDIA GeForce RTX 4090

The NVIDIA GeForce RTX 4090 is a flagship GPU for gaming and creation, built on the Ada Lovelace architecture with 24GB of GDDR6X memory. It delivers extreme performance for 4K and 8K gaming with support for DLSS 3 Frame Generation and full ray tracing.

GPU Architecture
Ada Lovelace
VRAM
2 4GB GDDR6X
CUDA Cores
16384
¥12,999¥14,999
SKU:NVIDIA-RTX4090-24G
In StockStock:99
NVIDIA Jetson Xavier NX Developer Kit
Jetson

NVIDIA Jetson Xavier NX Developer Kit

The NVIDIA Jetson Xavier NX Developer Kit delivers up to 21 TOPS of AI performance with 8GB of LPDDR4x memory, enabling powerful edge AI in an ultra-compact form factor. It supports multiple camera inputs for applications such as robotics, intelligent video analytics, and industrial inspection.

AI Compute
21 TOPS
memory
8GB LPDDR4x
GPU Architecture
Volta
¥4,999¥5,499
SKU:NVIDIA Jetson Xavier NX
In StockStock:99
NVIDIA GB200 NVL72
AI

NVIDIA GB200 NVL72

The NVIDIA GB200 NVL72 integrates 36 NVIDIA Grace CPUs and 72 NVIDIA Blackwell GPUs into a single rack-scale design. Powered by the NVIDIA NVLink domain and managed with NVIDIA Mission Control, it operates as one large GPU to deliver 30x faster inference for trillion-parameter large language models.

GPU Architecture
Blackwell
Number of GPUs
72
CPU
36 Grace CPUs
In StockStock:99
NVIDIA Jetson AGX Xavier Developer Kit
Jetson

NVIDIA Jetson AGX Xavier Developer Kit

The NVIDIA Jetson AGX Xavier Developer Kit is the previous-generation flagship embedded AI platform, delivering up to 32 TOPS of AI performance and featuring 32GB LPDDR4x memory. It supports multi-sensor fusion and complex AI workloads for advanced autonomous machines and edge computing applications.

AI Compute
32 TOPS
memory
32GB LPDDR4x
GPU Architecture
Volta
¥8,999¥9,999
SKU:NVIDIA JETSON AGX XAVIER
In StockStock:99

Need NVIDIA solutions?

As an NVIDIA partner, Huaying Technology provides end-to-end professional services—from selection consulting to deployment implementation.

Beijing ICP License No. 2026026760