Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Goal-conditioned Reinforcement Learning research benchmarks – Page 5 · SOTA2 Research

Goal-conditioned Reinforcement Learning

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

OGBench visual cube single play (5 tasks) zero-shot

14Average Return

4

Feb 26, 2026

DeepMind Control Suite P2E tasks standard (test)

968Walker Stand Score

3

Feb 26, 2026

69.2IQM (Normalized Win Rate)

2

May 14, 2026

95.5IQM (Normalised Win Rate)

2

May 14, 2026

95.6IQM

2

May 14, 2026

91.6IQM

2

May 14, 2026

99.5IQM (Success Rate)

2

May 14, 2026

Page 5 of 5

Popular datasets

OGBench Antmaze PointMaze humanoidmaze antsoccer

Follow for updates

Join SOTA2 Contact