a3124371940apache-2.0

RadeonVLA-Reflex Evaluation Videos

Public video and evidence companion for RadeonVLA-Reflex, a Physical AI submission containing evaluation videos, policy successes, and formal benchmark results for robotic fruit manipulation tasks.

Downloads0
Episodes100

Why This Matters for Physical AI

Demonstrates a vision-language-action model achieving 91% success on robotic fruit manipulation tasks through learned policies and explicit physics-based refinement, advancing embodied AI capabilities.

Technical Profile

Modalities
rgblanguage
Action Space
language
Environment
simulation
Task Types
manipulationpick_and_place
Episodes
100
Data Format
json
Annotation Types
language_instructionsaction_labels
License
apache-2.0
Part of the RadeonVLA-Reflex family

Access

Need custom rgb data?

Claru builds purpose-built datasets for simulation applications with dense human annotations and quality assurance.

Request a Sample Pack

Related Datasets