Name: MSHI - Embedded Edge Vision
Author: Heysuvi Labs

Question 1

What is Heysuvi Labs?

Accepted Answer

Heysuvi Labs is a Green AI lab building energy-efficient model systems for edge devices, private infrastructure, and local inference. Its two systems are MSHI for embedded edge vision and T-SUVI (SUVI 2.0) for multimodal AI.

Question 2

What is MSHI?

Accepted Answer

MSHI is an embedded edge vision system for factories, warehouses, service teams, field devices, cameras, robots, and OEM hardware. It is designed for settings where latency, power, and hardware limits define the deployment.

Question 3

Can MSHI run without a GPU?

Accepted Answer

Yes. MSHI is designed to run on ARM Cortex CPU cores. No discrete GPU is required in the deployment path.

Question 4

Can new classes be added to MSHI without retraining the model?

Accepted Answer

MSHI is designed for fast class updates without full model retraining. New objects and classes can be enrolled locally.

Question 5

What is T-SUVI (SUVI 2.0)?

Accepted Answer

T-SUVI, also called SUVI 2.0, is the new version of SUVI. It is a Green multimodal AI system that runs entirely on CPU, built for private workflows and efficient local inference on practical hardware.

Question 6

Can T-SUVI be used on-premises or in air-gapped environments?

Accepted Answer

Yes. T-SUVI is designed for on-prem infrastructure and air-gapped multimodal workflows, so private requests do not need to depend on external cloud APIs. A public release on Hugging Face is planned.

Question 7

Are MSHI and T-SUVI patented?

Accepted Answer

Patent applications covering the underlying model architecture have been filed in Ireland and the United States.

Question 8

Where is Heysuvi Labs based and how can I collaborate?

Accepted Answer

Heysuvi Labs, LLC has its US operations in Orlando, Florida. Enterprises, OEMs, and research groups can use the Collaborate button on this page or write to support@heysuvi.com.

Heysuvi Labs | Green AI Models for Edge and On-Prem Deployment