# Model deployment Grove Vision AI V2 can be deployed via an online website, which itself has a rich resource of models for you to choose from, so in addition to training your own models you can also use the models directly from the website. {% embed url="" %}

At the top of the page, you will see three columns titled: AI Models, Device Workspace, and About SenseCraft AI. click on Device Workspace and select Grove - Vision AI V2.

Click on the serial port you want to pair with, COM3 in the picture.

In the Models area, you can select models provided by SenseCraft AI or upload your own trained models.

Click ‘Select Model’ to select your favourite model.

It takes a little while to upload the model, so you will need to be patient.

In the preview section, you can observe the output results of the Grove Vision V2 's inference in real time, for example, it successfully detects cats after downloading the pet detection function.

In the ‘Device Logger’ section, the device sends real-time inspection data to the computer, and the data in the ‘Boxes’ section is： ``` { x; // 盒子中心的水平坐标 y; // 盒子中心的垂直坐标 w; // 识别框的宽度 h; // 识别框的高度 score; // 识别为目标的可信度 target; // 目标 } ``` Communicating with Petoi robot dogs using Grove Vision AI V2 You can use the Arduino IDE to modify our open source programme to use Grove Vision AI V2. Our programme integrates target tracking with Grove Vision AI V2. You only need to modify the code to enable this feature. For details on how to use it, see: [https://app.gitbook.com/o/-M-\_eWZUjFA4usjshHcZ/s/ntUipGYjnJE9HQWSovJT/](broken://spaces/ntUipGYjnJE9HQWSovJT). In addition, you can develop richer functionality using the APIs associated with the SSCMA library. --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://docs.petoi.com/extensible-modules/advanced-development-and-application-of-ai-vision-modules/model-deployment.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.