Deployment Routines

Artificial Intelligence Algorithm Deployment Routine

AIBOX-1688X has comprehensive capabilities for local private deployment of artificial intelligence algorithms. Whether it's a novel, high-computational-demand large language model or the classic YOLOv5 object detection model, AIBOX-1688X can handle both.

This chapter mainly introduces the SOPHON-DEMO project. For large model deployment, please refer to the "Large Model Deployment Routine" section.

SOPHON-DEMO Project Introduction

SOPHON-DEMO includes a series of transplantation routines for mainstream artificial intelligence algorithms, as well as detailed deployment documentation to facilitate user operation. Project repository: SOPHON-DEMO.

Routine List

The examples provided by SOPHON-DEMO are categorized from easy to difficult into three modules: tutorial, sample, and application:

• The tutorial module contains examples of basic interface usage.
• The sample module contains serial examples of classic algorithms on SOPHONSDK.
• The application module contains typical applications for typical scenarios.

| sample | Algorithm Category | Programming Language | BModel | |--- |--- |--- |--- | --- | | LPRNet | Licence Plate Recognition | C++/Python | FP32/FP16/INT8 | | ResNet | Image Classification | C++/Python | FP32/FP16/INT8 | | RetinaFace | Face Detection | C++/Python | FP32 | | segformer | Semantic Segmentation | C++/Python | FP32/FP16 | | SAM | Semantic Segmentation | Python | FP32/FP16 | | yolact | Instance Segmentation | C++/Python | FP32/FP16/INT8 | | YOLOv8_seg | Example splitting | C++/Python | FP32/FP16/INT8 | | PP-OCR | OCR | C++/Python | FP32/FP16 | | OpenPose | Human Body Keypoint Detection | C++/Python | FP32/FP16/INT8 | | C3D | Video Motion Recognition | C++/Python | FP32/FP16/INT8 | | DeepSORT | Multi-target Tracking | C++/Python | FP32/FP16/INT8 | | ByteTrack | Multi-target Tracking | C++/Python | FP32/FP16/INT8 | | CenterNet | Target Detection, Attitude Recognition | C++/Python | FP32/FP16/INT8 | | YOLOv5 | Target Detection | C++/Python | FP32/FP16/INT8 | | YOLOv34 | Target Detection | C++/Python | FP32/INT8 | | YOLOX | Target Detection | C++/Python | FP32/INT8 | | SSD | Target Detection | C++/Python | FP32/INT8 | | YOLOv7 | Target Detection | C++/Python | FP32/FP16/INT8 | | YOLOv8_det | Target Detection | C++/Python | FP32/FP16/INT8 | | YOLOv5_opt | Target Detection | C++/Python | FP32/FP16/INT8 | | ppYOLOv3 | Target Detection | C++/Python | FP32/FP16/INT8 | | ppYoloe | Target Detection | C++/Python | FP32/FP16 | | WeNet | Speech Recognition | C++/Python | FP32/FP16 | | BERT | Language Modelling | C++/Python | FP32/FP16 | | ChatGLM2 | Language Models | C++/Python | FP16/INT8/INT4 | | Llama2 | Language Models | C++/Python | FP16/INT8/INT4 | | ChatGLM3 | Language Models | Python | FP16/INT8/INT4 | | Qwen | Language Models | Python | FP16/INT8/INT4 | | Qwen1_5 | Language Models | Python | FP16/INT8/INT4 | StableDiffusionV1.5 | StableDiffusionV1.5 | Image Generation | Python | FP32/ FP16 | | GroundingDINO | GroundingDINO | Multimodal Target Detection | Python | FP16 | Applications | Application Scenarios