T-Chip Group
Intelligent hardware website
Innovate intelligent hardware and solutions.
Server Business Website
Server products and efficient solutions.
Consumer Business Website
StationPC NAS products and services.
CSC2-N48 Overview Specifications Wiki Download

CSC2-N48 Computing Power Server

Built-in 48 computing nodes

Private Deployment of Large Models

Configure aBMC Management System

Scalable M.2 PCIe NVMe SSD

Built-in 48 Computing Nodes

Server integrates 48 distributed computing nodes supporting multiple processor platforms (Rockchip / SOPHGO / NVIDIA) , with each node delivering 6~157 TOPS computing power. This scalable configuration provides powerful AI and deep learning acceleration capabilities.

Supports Private Deployment of Mainstream Large Models

Enables on-premises deployment of leading large language models (Deepseek-R1, Llama, Qwen), vision models (SAM, ViT), and image generation models (Flux, Stable Diffusion).

Supports Multiple Deep Learning Frameworks

Compatible with traditional network architectures (CNN, RNN, LSTM) and mainstream deep learning frameworks including TensorFlow, PyTorch, PaddlePaddle, and ONNX. Supports custom operator development and Docker container management technology.

Scalable M.2 PCIe NVMe SSD

Features 48 onboard M.2 PCIe interfaces supporting NVMe 2280 SSD expansion, enabling massive high-speed storage capacity. Delivers significantly improved data throughput to meet high-density storage requirements, with scalable capacity up to TB-level.

Multi-Software Smooth Operation

Seamlessly runs various software including: Databases (MySQL, Oracle、Redis), Web services (Nginx, Apache), Virtualization (KVM), Mail services (Postfix), Data analytics (Hadoop, Spark). Maintains stable and efficient performance during concurrent multi-software operations.

Touch Display for Real-time Monitoring

The device is equipped with a touchscreen display that provides real-time monitoring of chassis temperature, power efficiency, fan speed, network IP, date, time and other system parameters, allowing users to keep track of the operational status at all times.

Integrated Management System

Includes aBMC intelligent management system for: Real-time monitoring, Software/hardware configuration, Troubleshooting, Alert notification, System upgrades, Remote maintenance, Supports secondary development for customization.

Comprehensive Expansion Interfaces

Wide-ranging Application Scenarios

Widely applicable to intelligent computing servers, edge computing, large-scale model localization, smart cities, smart healthcare, smart industrial systems, intelligent security, and other related products and fields.

Edge Computing
Large-scale Model Localization
Smart Cities
Smart Healthcare
Smart Industrial Systems
Intelligent Security

Specifications

CSC2-N48
Technical Specifications Core Mods (Optional)

RK3588S

RK3588

RK35876

BM1684X

BM1688

Jetson Orin Nano ( 8GB )

Jetson Orin NX ( 16GB )
Server form

2U rackmount computing power server
Architecture

ARM architecture
Number of nodes

48 distributed computing nodes + 1 control node
Compute nodes

Octa-core 64-bit processor RK3588S, up to 2.4GHz

Octa-core 64-bit processor RK3588, up to 2.4GHz

Octa-core 64-bit processor RK3576, up to 2.2GHz

Octa-core 64-bit processor BM1684X, up to 2.3GHz

Octa-core 64-bit processor BM1688, up to 1.6GHz

Hexa-core 64-bit processor NVIDIA Jetson Orin Nano, up to 1.7GHz

Octa-core 64-bit processor NVIDIA Jetson Orin NX, up to 2.0GHz
Control nodes

Octa-core 64-bit processor RK3588, with a maximum clock speed of 2.4GHz and a peak computing power of 6TOPS
AI computing power

288T
(6T × 48，INT8)

288T
(6T × 48，INT8)

288T
(6T × 48，INT8)

1536T
(32T × 48，INT8)

768T
(16T × 48，INT8)

3216T
(67T × 48，INT8)

7536T
(157T × 48，INT8)
RAM

16GB LPDDR5
× 48
( 4/8/16/32GB )

16GB LPDDR4
× 48
( 4/8/16/32GB )

8GB LPDDR4
× 48
( 4/8/16GB )

8GB LPDDR4
× 48
( 8/12/16GB )

8GB LPDDR4
× 48
( 4/8/16GB )

8GB LPDDR5
× 48

16GB LPDDR5
× 48
Storage

256GB eMMC × 48 ( 16/32/64/ 128/256GB )

256GB eMMC × 48 ( 16/32/64/ 128/256GB )

64GB eMMC × 48 ( 16/32/64/ 128/256GB )

32GB eMMC × 48 ( 32/64/128GB )

32GB eMMC × 48 ( 16/32/64/ 128/256GB )

Not
Storage expansion

2280 PCIe NVMe SSD × 48 (Optional)

256GB (2280 PCIe NVMe SSD, the server is internally assembled)
Power

2 AC redundant power supplies (Hot-swappable supported)
Screen

1 touch screen display
Fan module

12 high-speed cooling fans
Physical Specifications Size

724.0mm (L) × 430.0mm (W) × 88.8 mm (H)
Full weight

Net weight of the server: 23.1kg, total weight with packaging: 25.3kg
Environment

Operating Temperature: 0ºC ~ 35ºC, Storage Temperature: -40ºC ~ 60ºC, Operating Humidity: 5% ~ 80%RH (non-condensing)
Software Specifications BMC

Integrated BMC management system based on web management interface, supports monitoring, configuration, alarming, remote operation and maintenance, virtual replacement management, and provides CLI command line and Redfish and other tools to facilitate secondary development
Large language models

All models support the privatization of ultra-large-scale parametric models under the Transformer architecture, such as Deepseek-R1 series, Gemma series, Llama series, ChatGLM series, Qwen series, Phi series and other large language models
Visual large model

BM1684X: Support the privatization deployment of large visual models such as ViT, Grounding DINO, SAM, etc. Jetson Orin Nano/Jetson Orin NX: Supports the privatization deployment of large vision models such as EfficientVIT, NanoOWL, NanoSAM, SAM, TAM, etc.
AI Painting

BM1684X/Jetson Orin Nano/Jetson Orin NX: Support the private deployment of Flux, Stable Diffusion, and Stable Diffusion XL image generation models
Deep learning

All models: Supports traditional network architectures such as CNN, RNN, LSTM, and supports a variety of deep learning frameworks, including TensorFlow, PyTorch, PaddlePaddle, ONNX, Caffe, etc. Support custom operator development and Docker containerization management technology Jetson Orin Nano/Jetson Orin NX: Support Ollama local large model deployment framework and ComfyUI graphical deployment framework
Interface Specifications Internet

4 × 10G Ethernet (SFP+), 1 × Gigabit Ethernet (RJ45, MGMT used as BMC management network)
Console

1 × Console (RJ45, BMC debug serial port, baud rate 115200)
Display

1 × HDMI (Maximum resolution 1080P, BMC management display)
USB

2 × USB3.0 (The lower USB is USB3.0 OTG, and the BMC can be upgraded OTG using a USB flash drive)
Button

1 × Reset button, 1 × Power button, 1 × Restart BMC button