avanturist2025MIT

MIKASA-Robo

A comprehensive benchmark suite of memory-intensive robotic manipulation tasks with ready-made datasets and pre-trained oracle agent checkpoints for reinforcement learning research.

Downloads605
Likes6

Why This Matters for Physical AI

MIKASA-Robo provides a memory-focused benchmark for evaluating reinforcement learning agents on complex robotic manipulation tasks that require temporal reasoning and state memory.

Technical Profile

Environment
tabletopsimulation
Task Types
manipulationreinforcement-learning
License
MIT
Part of the MIKASA-Robo family

Community Signals

Top 50% by downloads

Access

Need custom physical AI data?

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

Request a Sample Pack