Skip to content
Sujan Ghimire
Research
Exploratory

Vision-Language-Action Models on Diverse Computing Platforms

Exploratory study of how vision-language-action models deploy and perform on different computing platforms.

  • AI/ML

Problem

Vision-language-action (VLA) models connect perception and language to physical actions, but their computational demands raise questions about how they behave when deployed on different hardware platforms.

Technical approach

  1. 01Study deployment of VLA models across different computing platforms.
  2. 02Characterize their performance behaviour on each platform.

Associated publications

This is exploratory work; no publications are associated with it yet.