1. Core product information

product name: TIM-SN300 multi-modal identification and diagnosis AI large model all-in-one machinebrand: TETRN BOT Tianchuang Robotproduct positioning: Edge analysis equipment for industrial scenarios that deeply integrates hardware computing base and AI algorithm platformcore label: Standard system platform, built-in AI algorithm capabilities, covering domestic and foreign mainstream AI models

2. AI algorithm capabilities

core value: Help enterprise customers build high-performance AI algorithm applications

1. Algorithm free call control: Support AI algorithm selection and customization of model thresholds and alarm parameters.
2. algorithm self-optimizing: Based on the running data samples, the algorithm system adjusts and optimizes itself to adapt to the scene recognition target.
3. algorithm supply: 5 categories of 100+AI identification algorithms, covering equipment, facilities, personnel, and environment identification and testing.
4. Core identification scenarios: Equipment status assessment, material defect identification, environmental status monitoring, personnel behavior identification, instrument data collection.
5. Algorithm library support: Configure more than 100 algorithm libraries in five categories: meter status reading, scene environment monitoring, structural defect identification, equipment health management, and personnel behavior supervision, which are called on demand and freely combined.

3. System platform capabilities

core value: Quickly build a customized management and control system platform for corporate customers

1. Core Capability Module: Capability base, equipment IoT, data access, diagnostic analysis.
2. Internet of Everything: Supports simultaneous connection of multiple types of sensing hardware, allowing smooth connection with third-party platforms.
3. data collaboration: Real-time data collection and transmission, supporting the access and integration of multiple and heterogeneous data sources.
4. Diagnostic Analysis: Based on self-developed large-scale model cloud computing, realize statistics, diagnosis, and analysis of monitoring data, and build a visual display of multi-dimensional data trends.
5. Platform functional interface: Algorithm configuration, information overview, real-time detection item configuration, real-time monitoring, task management, data statistics.

4. Hardware core configuration

1. core processor: Equipped with ThinkForce TF7 XXA series AI processor
2. storage configuration
   • Factory pre-installed 3.5-inch SATA data disk
   • DDR44 memory: 32GB/64GB
   • System disk SSD: 500GB
   • Data disk: 4T/8T
3. interface configuration: 2 PCIe expansion interfaces, 4 USB3.0+2 USB2.0 interfaces, 2 network interfaces, 1 GbE system management network interface
4. deployment features: Out of the box, flexible deployment; support rack installation, software and hardware integration.

V. Key technologies and characteristics

1. Large model adaptation: Based on vertical fine-tuning of Qwen, Deepseek, O1, World VLA and other large models, it adapts to all current mainstream open source large model platforms.
2. independent and controllable: Comprehensively use localized firmware to build the hardware device layer, and the system software layer supported by the database and operating system is completely autonomous and controllable.
3. system platform: Tianxuan AI perception and identification diagnosis system platform (integrated system integrating algorithm deployment, management, and configuration).

6. Application scenarios

One-stop operation and maintenance services, real-time monitoring of equipment on-site operation and equipment health management throughout the life cycle (suitable for industrial production lines, energy and chemical industry and other industrial scenarios).