Certifications:
| PRODUCT | T906G EDGE Computer 32GB | T906G EDGE Computer 64GB | ||
| AI Computing power | 200 TOPS | 275 TOPS | ||
| GPU | 1792core NVIDIA Ampere architecture GPU with 56 Tensor Cores 930MHz | 2048core NVIDIA Ampere GPU with 64 Tensor Cores1.3 GHz | ||
| CPU | 8-core Arm® Cortex®-A78AE v8.2 64-bit CPU 2MB L2 + 4MB L3 2.2GHz | 12-core NVIDIA Arm® Cortex A78AE v8.2 64-bit CPU3MB L2 + 6MB L32.2GHz | ||
| Memory | 32GB256-bit LPDDR5 204.8GB/s | 64GB 256-bit LPDDR5204.8 GB/s | ||
| System storage | 128GB SSD Standard config | |||
| video coding | 1x 4K60 | 3x 4K30 | 6x 1080p60 | 12x 1080p30 (H.265) 1x 4K60 | 2x 4K30 | 5x 1080p60 | 11x 1080p30 (H.264) | |||
| Video decoding | 1x 8K30 | 2x 4K60 | 4x 4K30 | 9x 1080p60| 18x 1080p30 (H.265) 1x 4K60 | 2x 4K30 | 5x 1080p60 | 11x 1080p30 (H.264) | |||
| External interface | ||||
| Ethernet | 1xRJ45 1000 BASE-T 1xRJ45 1*10Gbits | |||
| Wireless network | 2.4GHz/5GHz Dual-band WiFi | |||
| SIM Card | 4G/5G Full Netcom丨nano sim card slot | |||
| USB | 4×USB 3.0/2.0TYPE A丨 1xOTG TYPE C | |||
| Camera | 8 GMSL1/GMSL2 | |||
| Video output | 1xHDMI 2.1 TYPE A | |||
| CAN | 2 xCAN 2.0b | |||
| Other I/O | 2xGPIO(in\put)（1*3.3VTTL，1*1.8VTTL） | |||
| UART | 3x RS232 (compatible with RS485) | |||
| M.2 Port | 1xM.2 Key M（PCIE NVME 2280 SSD） | |||
| Power input | Wide input 12-36V DC | |||
| Size | 196.7mm x 196mm x 74mm | |||
1. 12 months warranty. For quality problem, we will send you a new one.
2. Technical support. We provide online one-to-one technical support to help you shorten your development cycle.
3.Pre-installation of drivers. We pre-install drivers of many kind of devices for saving you the trouble of driver development.
The core module of T906G is NVIDIA Jetson AGX Orin (32GB/64GB), smaller than a credit card. This mass-produced module system (SOM) can provide strong support for deploying AI on edge devices in multiple industries, including smart cities, smart factories, agriculture, and robotics.
| Jetson AGX Orin 64GB | Jetson AGX Orin 32GB | |
| AI Performance | 275 TOPS | 200 TOPS |
| GPU | A 2048-core NVIDIA Ampere architecture GPU with 64 Tensor cores | A 1792-core NVIDIA Ampere c GPU with 56 Tensor cores |
| CPU | 12 core Arm® Cortex®-A78AE v8.2 64 bit CPU 3MB L2 + 6MB L3 | 8 core Arm® Cortex®-A78AE v8.2 64 bit CPU 2MB L2 + 4MB L3 |
| Memory | 64GB 256 bit LPDDR5 204.8GB/s | 32GB 256 bit LPDDR5 204.8GB/s |
| Storage | 64GB eMMC 5.1 | 64GB eMMC 5.1 |
| Video Encode | 2x 4K60 (H.265) 4x 4K30 (H.265) 8x 1080p60 (H.265) 16x 1080p30 (H.265) | 1x 4K60 (H.265) 3x 4K30 (H.265) 6x 1080p60 (H.265) 12x 1080p30 (H.265) |
| Video Decode | 1x 8K30 (H.265) 3x 4K60 (H.265) 7x 4K30 (H.265) 11x 1080p60 (H.265) 22x 1080p30 (H.265) | 1x 8K30 (H.265) 2x 4K60 (H.265) 4x 4K30 (H.265) 9x 1080p60 (H.265) 18x 1080p30 (H.265) |
| CSI Camera | Up to 6 cameras(16 supported via virtual channels) 16-channel MIPI CSI-2 D-PHY 2.1 (up to 40 Gbps) | C-PHY 2.0 (up to 164 Gbps) | Up to 6 cameras (16 supported via virtual channels) 16-channel MIPI CSI-2 D-PHY 2.1 (up to 40 Gbps) | C-PHY 2.0 (up to 164 Gbps) |
| PCIe | Up to 2 x8 + 1 x4 + 2 x1 （PCIe 4.0、Root port and endpoint） | Up to 2 x8 + 1 x4 + 2 x1 （PCIe 4.0、Root port and endpoint） |
| USB | 3x USB 3.2 2.0 (10 Gbps) 4x USB 2.0 | |
| Network | 1x GbE 1x 10GbE | |
| Display | 1x 8K60 multi-mode DP 1.4a (+MST)/eDP 1.4a/HDMI 2.1 | |
| Other I/O | 4x UART、3x SPI、4x I2S、8x I2C、2x CAN、PWM、DMIC and DSPK、GPIO | |
| Power consumption | 15 W – 60 W | 15 W – 40 W |
Application Fields:
Support multi scenario applications,
support parallel operation of 5 algorithms for 16 video streams
1 Practical Application in Security Monitoring Scenarios
2 Practical Application in Smart Gas Station
Practical Application in Smart Petroleum
Simultaneously supporting channel video analysis
Support smoke, fire, personnel area monitoring, personnel fall,
safety helmet recognition, tooling, gauges, oil level, valve direction
4 Practical Application | Identification of unidentified Objects in Power Transmission
Simultaneously supporting 4-channel video analysis
Unidentified object recognition, fireworks, engineering vehicles
Subsequent additions, abnormal insulators/pins/bird spikes, etc
If you need any customized services, please feel free to contact us.
Highly professional: we’ve been focusing on edge computing & embedded harware since 2011.
Fast delivery time: due to the mature supply chain of Shenzhen, our on-time delivery rate can up to 99% even during the stage of chip shortage.
Guranteed quality: comply with the industrial standard, passed all the important tests like FCC, CE, 3C, etc.
