Questions We Hear First.

Yes, at selected sites. AI Vision Bots is offered by enquiry, starting with sites where the VYROX AI team can work with you on cameras, models and rules. It is not sold as a boxed product.

It depends on the site, the cameras and the Detections. There is no price list; send an enquiry and we will scope it with you.

ONVIF cameras can be found by search, and any camera with an RTSP address can be added. TCP, UDP and auto transport are supported, with Basic, MD5 Digest or SHA-256 Digest sign in.

It does not have to. Detections and recording can run on the AI Vision Bot at the site, sending only events, snapshots and counts. Clips can stay on site and are fetched only when someone asks for them.

It is built to. AI Vision Bots records 24/7 or by event, schedule or motion, keeps clips with pre and post roll, and offers timeline playback and MP4 export.

A Mac, Linux or Windows computer at the site, x86-64 or ARM64. A GPU is optional and makes Detections faster.

Yes, in the AiServa AI Vision Models app on your own GPU server. Trained models run as Detections on your cameras.

Yes. The AI assistant turns a plain language request into a rule with its jobs. A person confirms before it is saved.

Our own VYROX Vision Engine, plus models you train in AI Vision Models. Each use case uses only the engines it needs, retrained on footage from your site.

A computer vision service from VYROX AI in Malaysia. We pick the use case with you, train on your own footage, install on site and stay on to tune it, for factories, farms, quarries, roads, shops and safety teams.

Usually not. Most IP and CCTV cameras at 1080p work. Small objects, fine defects, underwater views and long range need the right lens, light and mounting, which we check on site.

No. The 3D demos are simulations made to explain each use case. On a real project we train and test models on your own footage and share the accuracy before you commit.

The computer at your site that is paired to AiServa. It connects to your cameras, runs Detections, records, and takes updates from AiServa. AI Vision Bots is the app you use to manage it.

No. Live view is built to send video over WebRTC straight from the site to the viewer. A relay is used only when a direct path is not possible.

Terminal and Remote Desktop are built with an off switch on the AI Vision Bot computer, and each session is recorded in an audit log. The switch on site always wins.

A Detection is a model with its classes and thresholds running on one camera. For example, a model with the classes truck, person and machine can run on a gate camera. The classes are examples; you choose your own.

You choose Add AI Vision Bot in AiServa and run one short command on the computer at your site, the same way you add a server. The code works once and expires in minutes. A six digit approval code on the computer is then confirmed on AiServa.

The AI Vision Bot keeps running from its last signed configuration. Changes made on the AI Vision Bot and events wait in a queue and are sent in order when it reconnects.

If different fields were changed, both changes are kept. If the same field was changed on both sides while apart, the later save wins, the earlier value stays in history, and the person whose change was replaced gets a notice with a way to restore it.

Nobody in readable form. Camera passwords are sealed to the AI Vision Bot, are never shown in the app, and do not appear in a Terminal session.

Yes. Other systems, such as IoTServa and AI Quarry, receive Detections, events, snapshots and clips from AI Vision Bots through AiServa, set up together with the VYROX team.

Send a 30 to 60 second clip from the camera you would use, taken at a normal working time, and one sentence on the problem. A phone recording of the screen is fine to begin with.