AI Inference

AI Nvidia Jetson

Transforming Data, Shaping the Future

AI inference is the crucial step in which our trained models apply their acquired knowledge to make precise decisions or generate predictions based on new data. By training our AI models on extensive datasets, they gain the ability to recognize patterns and process information. During the inference phase, these models use the knowledge they’ve learned to generate relevant insights or solve specific problems. This process enables our AI to respond to queries in real time and deliver well-informed decisions. This allows us to apply it in various practical ways.

Jetson Orin

High-performance edge AI inference computer with integrated NVIDIA Jetson architecture—ideal for demanding video applications in surveillance, transportation, and manufacturing.

MIC-733-AO

MIC-733_AO_Serveransicht_Links
AI System Based on NVIDIA® Jetson AGX Orin™

 

  • Compact, fanless design

  • NVIDIA® Jetson AGX Orin™

  • Up to 275 TOPS

  • 4x GbE (optional PoE) and 4x USB 3.2 Gen 2 (10 Gbit/s)

  • 2x mPCIe and 2x Nano-SIM slots

  • Allxon 24/7 remote monitoring and OTA deployment; Azure Certified Devices

MIC-713-OX

MIC-713_Serveransicht
AI System Based on NVIDIA® Jetson Orin™ NX

 

  • Compact, fanless design
  • NVIDIA® Jetson Orin™ NX
  • Up to 100 TOPS
  • 2x GbE and 6x USB 3.2 Gen 2
  • 1x mPCIe, 1x M.2 3052, and 1x M.2 2280
  • Allxon 24/7 remote monitoring and OTA deployment

MIC-713-ON

MIC-713_Serveransicht
AI Platform Based on NVIDIA® Jetson Orin™ Nano

 

  • Compact, fanless design
  • NVIDIA® Jetson Orin™ Nano
  • Up to 67 TOPS
  • 2x GbE and 6x USB 3.2 Gen 2
  • 1x mPCIe, 1x M.2 3052, and 1x M.2 2280
  • Allxon 24/7 remote monitoring and OTA deployment

MIC-712-OX

MIC-712_Serveransicht
AI Platform Based on NVIDIA® Jetson Orin™ NX

 

  • Ultra-slim design for mobile devices in security applications
  • NVIDIA® Jetson Orin™ NX / Orin Nano™
  • Up to 157 TOPS in NVIDIA Super Mode
  • Expansion options for GMSL, PoE, and USB
  • Integrated out-of-band management
Data Sheet
Data Sheet
Data Sheet
Data Sheet

MIC-712-ON

MIC-712_Serveransicht
AI Platform Based on NVIDIA® Jetson Orin™ Nano

 

  • Ultra-slim design for mobile devices in security applications
  • NVIDIA® Jetson Orin™ Nano
  • Up to 67 TOPS in NVIDIA Super Mode
  • Expansion options for GMSL, PoE, and USB
  • Integrated out-of-band management

MIC-712P-OX

MIC-712_Serveransicht
AI Platform Based on NVIDIA® Jetson Orin™ NX

 

  • Ultra-slim design for mobile devices in security applications

  • NVIDIA® Jetson Orin™ NX

  • Up to 157 TOPS in NVIDIA Super Mode

  • 6x PoE ports, each providing 25 W of power

  • Integrated out-of-band management

MIC-712P-ON

MIC_712P_Serveransicht_links
AI Platform Based on NVIDIA® Jetson Orin™ Nano

 

  • Ultra-slim design for mobile devices in security applications

  • NVIDIA® Jetson Orin™ Nano

  • Up to 67 TOPS in NVIDIA Super Mode

  • 6x PoE ports, each with 25 W power supply

  • Integrated out-of-band management

MIC-711-OX

MIC-711_Serveransicht_links2
AI Inference System Based on NVIDIA Jetson Orin NX

 

  • Fanless, ultra-compact design

  • NVIDIA® Jetson Orin™ NX

  • Up to 157 TOPS

  • 1x GbE and 2x USB 3.2 Gen 1 (5 Gbit/s)

  • 1x mPCIe, 1x M.2 3052, and 1x M.2 2280

  • Allxon 24/7 remote monitoring and OTA deployment; Super Mode available starting with JetPack 6.2

Data Sheet
Data Sheet
Data Sheet
Data Sheet

MIC-711-ON

mic-711-on1_Serveransicht_links
AI Inference System Based on NVIDIA® Jetson Orin™ Nano

 

  • Fanless, ultra-compact design

  • NVIDIA® Jetson Orin™ Nano

  • Up to 67 TOPS

  • 1x GbE and 2x USB 3.2 Gen 1 (5 Gbit/s)

  • 1x mPCIe, 1x M.2 3052, and 1x M.2 2280

  • Allxon 24/7 remote monitoring and OTA deployment; Super Mode available starting with JetPack 6.2

MIC-733-TS

MIC_733_TS_Serveransicht
NEMA-TS2-KI inference system based on NVIDIA® Jetson AGX Orin™

 

  • Compact, fanless design; operating temperature range from -34 to 74 °C

  • NVIDIA® Jetson AGX Orin™

  • 500-ms power interruption recovery in accordance with NEMA TS2

  • NEMA TS2-compliant

MIC-715-OX

MIC-715_OX_Serveransicht_links
Rugged, fanless vehicle AI system based on NVIDIA® Jetson Orin™ NX

 

  • Rugged, waterproof, and fanless design compliant with IP67

  • NVIDIA® Jetson Orin™ NX

  • Up to 100 TOPS

  • 2x GbE, 4x PoE compliant with IEEE 802.3af, 2x USB 3.0, 1x CAN bus, and 1x HDMI

  • 2x mPCIe, 1x M.2 3052, and 1x M.2 2280

  • Allxon 24/7 remote monitoring and OTA deployment

MIC-736

MIC_736_Serveransicht_links
AI system based on NVIDIA Dual AGX ORIN

 

  • NVIDIA Dual AGX ORIN with a 12-core NVIDIA Arm® Cortex-A78AE CPU and an Ampere-based GPU
    with 2,048 CUDA cores and 64 Tensor cores

  • 64 GB LPDDR5 and 64 GB eMMC

  • Optimized for robotics and mobility AI applications

Data Sheet
Data Sheet
Data Sheet
Data Sheet

Jetson Thor

Next-generation, high-performance edge AI for cutting-edge robotics, industrial automation, and autonomous systems.

MIC-741-AT

MIC_741_AT_Serveransicht_Links
AI inference system based on NVIDIA® Jetson Thor™

 

  • NVIDIA® Jetson T4000™ with up to 1,200 TFLOPS (FP4)
  • 1x SFP28 (25 GbE), 2x GbE, 2x 10GbE, and 6x USB 3.2 Gen 2
  • 2x M.2 2280, 1x M.2 E-Key for Wi-Fi, and 1x M.2 B-Key for LTE/5G
  • 2x isolated CAN FD (8 Mbit/s)
  • Up to 8 GMSL 2.0 channels with Mini-FAKRA connectors
  • Designed for generative AI applications with large AI models

MIB-741-AT

MIB_741_AT
AI Inference Platform Based on NVIDIA® Jetson Thor™

 

  • NVIDIA® Jetson T5000™ / Jetson T4000™ with up to 2,070 TFLOPS (FP4)

  • 1x SFP28 (25 GbE), 2x GbE, 2x 10GbE, and 6x USB 3.2 Gen 2

  • 2x M.2 2280, 1x M.2 E-Key for Wi-Fi, and 1x M.2 B-Key for LTE/5G

  • 4x isolated CAN FD (8 Mbit/s)

  • Up to 8 GMSL 2.0 channels with Mini-FAKRA connectors

  • JetPack 7.1; variants with T5000 (128 GB) or T4000 (64 GB LPDDR5X)

MIB-742-AT

MIB-742_3
AI Inference SBC based on NVIDIA® Jetson Thor™

 

  • NVIDIA® Jetson T5000™ / Jetson T4000™ with up to 2,070 TFLOPS (FP4)

  • 1x QSFP28 with up to 4x 25GbE (T5000) or 3x 25GbE (T4000)

  • 1x 5GbE and 4x USB 3.2 Gen 2

  • 1x M.2 E-Key for Wi-Fi and 1x M.2 B-Key for LTE

  • 4x CAN FD (on T5000), 1x I2C, and up to 8 channels of GMSL 2.0 with Mini-FAKRA

  • JetPack 7.0; variants with T5000 (128 GB) or T4000 (64 GB LPDDR5X)

MIB-735

MIB_735_Frontalansicht
AI Inference Board Kit Based on NVIDIA IGX Thor™

 

  • NVIDIA IGX™ T5000 with up to 2,070 FP4 TFLOPS and 128 GB LPDDR5X

  • 4x 25GbE via SFP28 and 1x 5GbE Ethernet

  • Functional safety MCU for safety-critical applications

  • Rugged design for an operating temperature range of -30 °C to +70 °C

  • Holoscan Sensor Bridge ready for robotics and sensor applications

  • Optional up to 8 GMSL 2.0 channels with FAKRA connectors

Data Sheet
Data Sheet
Data Sheet
Data Sheet

MIC-743-AT

MIC_743_AT_Serveransicht
AI Inference System Accelerated by NVIDIA® Jetson Thor™

 

  • NVIDIA® Jetson T5000™ / Jetson T4000™
  • Up to 2070 TFLOPS (FP4)
  • 1x QSFP28 (T5000: 4x 25GbE / T4000: 3x 25GbE), 1x 5GbE, and 4x USB 3.2 Gen 2
  • 1x M.2 AE-Key (Wi-Fi) and 1x M.2 B-Key (LTE)
  • Designed for versatile generative AI applications with large AI models

MIC-742-AT

MIC_742_AT_Serveransicht_links
AI Inference System Accelerated by NVIDIA® Jetson Thor™

 

  • NVIDIA® Jetson T5000™

  • Up to 2,070 TFLOPS (FP4)

  • 1x QSFP28 (T5000: 4x 25GbE), 1x 5GbE, 4x USB 3.2 Gen 2, 4x CAN bus, 1x I2C, and 8 channels of GMSL 2.0 with Mini-FAKRA connectors

  • 1x M.2 AE-Key (Wi-Fi) and 1x M.2 B-Key (LTE)

MIC-735-IT

MIC-735_IT_Serveransicht_Links
AI inference system based on NVIDIA IGX Thor™

 

  • Rugged design for an operating temperature range of -30 °C to +60 °C

  • NVIDIA IGX™ T5000

  • Up to 2,070 FP4 TFLOPS

  • 4x 10G SFP and 1x 5G Ethernet

  • Functional safety MCU and Holoscan Sensor Bridge ready for robotics applications

Data Sheet
Data Sheet
Data Sheet

Jetson for Automotive Applications

Rugged, fanless AI inference systems for use in railways, rail vehicles, and commercial vehicles, based on NVIDIA Jetson.

ITA-510

ITA_510_Serveransicht_links
AI Inference System Based on NVIDIA® Jetson Orin™ NX / Orin™ Nano for Rail Applications

 

  • Fanless system for AI inference applications

  • NVIDIA® Jetson Orin™ NX / Orin™ Nano

  • Rugged M12 connectors for harsh industrial environments

  • 1x Mini PCIe, 1x M.2 (E-Key), 1x M.2 (M-Key), and 1x M.2 (B-Key)

  • Compliant with EN 50155 OT1 and the EMC requirements of EN 50121-3-2

ITA-560AGX

ITA_560AGX_Serveransicht_Links
AI Inference Box PC Based on NVIDIA® Jetson AGX Orin™ for Rail Applications

 

  • Fanless Box PC for AI Inference Applications

  • NVIDIA® Jetson AGX Orin™

  • Rugged M12 connectors for harsh industrial environments

  • 1x Mini PCIe and 1x M.2 (M-Key)

  • Compliant with EN 50155 OT1 and the EMC requirements of EN 50121-3-2

ITA-560

ITA_560_Serveransicht_links
AI Inference Box PC Based on NVIDIA® Jetson Orin™ NX / Orin™ Nano for Rail Applications

 

  • Fanless Box PC for AI Inference Applications
  • NVIDIA® Jetson Orin™ NX / Orin™ Nano
  • Rugged M12 connectors for harsh industrial environments
  • 1x Mini PCIe, 1x M.2 (E-Key), 1x M.2 (M-Key), and 1x M.2 (B-Key)
  • Compliant with EN 50155 OT1 and the EMC requirements of EN 50121-3-2
Data Sheet
Data Sheet
Data Sheet

Jetson for Medical Applications

Compact medical systems with IPX1-certified enclosures—perfect for space-saving and safe use in clinics and hospitals.

USM-500 NVIDIA IGX Orin

USM-500-Serveransicht
Professional Tower/Standalone Medical Computer with NVIDIA IGX

 

  • Stylish and user-friendly tower/standalone design for hospital environments

  • NVIDIA IGX with advanced functional safety and security features

  • Up to 275 TOPS

  • NVIDIA® ConnectX®-7 SmartNIC with up to 200 Gbit/s network performance

  • Designed for medical AI applications and the secure capture of high-frame-rate raw video data

USM-500 NVIDIA IGX Thor

USM-500-Serveransicht
Professional tower/standalone medical computer with NVIDIA IGX Thor™

 

  • Tower/Standalone System for Business-Critical Medical Environments

  • NVIDIA IGX Thor™ with functional safety and security

  • Up to 5,581 FP4 TFLOPs (sparse) with NVIDIA RTX PRO™ 6000 Blackwell Max-Q Workstation Edition

  • Integrated dual 100GbE NVIDIA ConnectX® 7 network

  • PCIe Gen5 and 128 GB LPDDR5X memory

  • Supports IEC/EN 60601-compliant integration of medical systems

data sheet
Data Sheet

Request your AI inference system from Thomas-Krenn here!

No results found
No results found
No results found