
SXD-NH4 AI Video Stream Analytics Server (192 TOPS)
Specification · 192 TOPS@INT8 · 256 Channels + 1080P
Shixinda AI’s Flagship AI Video Analytics Server delivers peak computing power up to 192 TOPS, supports real-time analysis of over 256 HD video channels, and is…
No. sxd-nh4-ai-video-stream-analytics-server-192-tops · 2026-11-19
Overview
Product Overview
Shixinda AI SXD-NH4 is a flagship AI video analytics server with a peak compute power of up to 192 TOPS. Designed specifically for large city-level monitoring centers and ultra-large enterprise headquarters, it is built to handle massive video analytics workloads.
Core Features
- 192 TOPS peak compute power, supporting real-time analysis of over 256 channels of 1080P video
- 4U rack-mounted design, supports multi-GPU expansion
- Full redundant design with 99.99% availability
- Supports cluster deployment and horizontal scaling
Typical Applications
City-level Security Command Center, Provincial-level Monitoring Centers, Large Smart City Operation Hubs
Capacity and channel definition
The quoted channel count refers to live video channels under a baseline of 1080P, 25 fps, one algorithm per channel and one sampled frame every 2 seconds. Real capacity varies with three factors:
| Variable | Change | Effect on compute |
|---|---|---|
| Camera resolution | 2 MP → 4 MP | Decode and inference load per channel rises by 50%–100%; channel count drops |
| Sampling interval | 2 s → 1 s | Frames processed per second double; channel count roughly halves |
| Algorithms per channel | 1 → 2 | Inference calls double; channel count roughly halves |
Rule of thumb: convert by "resolution × sampling rate × algorithms per channel", then keep 30% headroom. 2 MP with 2-second sampling and a single algorithm corresponds to the quoted channel count; for 4 MP or 1-second sampling, plan for 50%–60% of it.
Dataflow architecture and effective compute
The nominal TOPS is only an upper bound — what determines real capacity is chip utilization. This device runs on a reconfigurable dataflow AI chip: the whole model is compiled into a data pipeline, so operands flow through the hardware and computation completes on the way, without repeatedly fetching instructions or moving intermediate results in and out of memory.

| Item | Instruction-set (typical GPU / NPU) | Dataflow (this device) |
|---|---|---|
| Execution | operators run one by one, intermediates round-trip through memory | operator graph compiled into a pipeline |
| Chip utilization (public comparison) | 10%–30% | 65%–95% |
| Effective compute at 8 TOPS | about 0.8–2.4 TOPS | about 5.2–7.6 TOPS |
| Channels at equal nominal TOPS | baseline | about 2–3× by median utilization |
| Model deployment | framework adaptation and manual tuning | in-house compiler, TensorFlow / PyTorch / ONNX compiled automatically |
Effective compute = peak TOPS × utilization. Three practical consequences: fewer devices for legacy upgrades (saving rack space and power), algorithms can be recompiled and redeployed without replacing hardware, and utilization stays high under concurrent decoding plus inference.
Utilization figures are publicly reported architecture ranges; measured values vary with operator structure, input resolution and batching strategy.
Algorithms
Algorithms are grouped into six scenario families, covering 460+ fine-grained scenarios. Which ones are enabled is selected per project and confirmed in the delivery list.
| Scenario family | Representative algorithms | Typical locations |
|---|---|---|
| People and attendance | intrusion, line crossing, loitering, lingering, climbing, fall, on-duty status, PPE compliance, helmet / vest / uniform / mask / glove detection | perimeter, guard room, work area |
| Safety and fire | smoke, open flame, fire lane obstruction, restricted-area intrusion, blocked passage, gas leak, static discharge | warehouse, switch room, kitchen, hazardous area |
| Vehicles and traffic | plate recognition, illegal parking, wrong-way driving, traffic counting, non-motor vehicle intrusion, engineering vehicles, height limit | campus gate, site entrance, road |
| Objects and facilities | material stacking, pallets, shelf gaps, equipment status, gauge reading, conveyor anomaly | workshop, warehouse, power room |
| Environment and order | bin overflow, standing water, dust, street occupation, clutter | site roads, waste room, construction area |
| Food and kitchen | kitchen uniform, rodent detection, sample retention, tableware sanitising, cold-chain temperature | school and staff canteens |
Alert data flow

Deployment architecture

Algorithm training & customization
Shixinda AI also provides an algorithm training platform: annotate site data, train segmentation / detection / classification models, evaluate, then publish a quantized SG file and load it onto the edge devices.

Related
- Product: Shixinda AI AI Accelerator Card (Edge Inference)
- Product: Shixinda AI SXD-N430-8 Edge Analytics Station (8 TOPS)
- Product: Shixinda AI SXD-X6A Edge Intelligence Analytics Box
- Solution: Smart Campus Security Monitoring Upgrade Solution
- Solution: Shixinda AI Smart Classroom Behavior Analytics Solution
References
Related Chinese regulations and standards for further reference:
- 《公共安全视频监控联网系统信息传输、交换、控制技术要求》(GB/T 28181) — official source
- 《中华人民共和国数据安全法》 — official source
Applications




Core Features
This section is pending content and will appear once filled in the admin.
This section is pending content and will appear once filled in the admin.
This section is pending content and will appear once filled in the admin.
This section is pending content and will appear once filled in the admin.
Specifications
Hardware Specifications
| Item | Specification |
|---|---|
| AI Computing Power | 192 TOPS@INT8 |
| Video Input | 256 Channels + 1080P |
| Form Factor | 4U Rack-mount |
| Power Supply | Redundant Power Supplies ×2 |
| Memory | 128GB ECC |
| Expansion | Multi-card Expansion, Cluster Deployment |
| Availability | 99.99% |
| Operating Temperature | 10℃~35℃ |
FAQ
Does SXD-NH4 work with our existing cameras?
Yes. It connects to existing IP cameras over RTSP or GB/T 28181 and keeps your current cameras and NVR, so no camera replacement is needed.
How many channels can SXD-NH4 handle?
The rated capacity is %s. Actual capacity changes with camera resolution, frame sampling interval and the number of algorithms per channel.
What is included in delivery?
Supply, on-site installation and commissioning, platform integration and operator training, with remote support afterwards.
Related Products
View All →
Shixinda AI AI Video Analytics Server
48 TOPSShixinda AI Legacy Cameras + AI Brain: High-performance Video Analysis Server deployed in monitoring centers.
Learn More →
Shixinda AI Video Analytics Server (96-192 TOPS Series)
192 TOPSOld Cameras + AI Brain: Upgrade Solution for Large-Scale Legacy Surveillance Peak computing power up to 192 TOPS, supports intelligent analytics for 64–512 vi…
Learn More →
SXD-NH1 Video Analytics Server
48 TOPSShixinda AI Standard Rack-Mount AI Video Analytics Server, delivers 48 TOPS peak computing power, supports real-time analytics for 64 channels of HD video, idea…
Learn More →
Shixinda AI SXD-NH2 Video Analytics Server
96 TOPSShixinda AI’s high-performance rack-mounted AI video analytics server delivers 96 TOPS of peak computing power, supports real-time analysis for 128 channels of…
Learn More →
← PrevShixinda AI SXD-NH2 Video Analytics Server (96 TOPS)Next →SXD-N430S Edge Analytics Appliance (8 TOPS)
Ready to go smart?
Leave a message or call us for tailored selection advice.




















