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

Continuous Control research benchmarks – Page 8 · SOTA2 Research

Continuous Control

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

Walker2d MuJoCo v2 (evaluation)

852.3Score (Action p=0.1)

7

Mar 10, 2026

HalfCheetah MuJoCo v2 (evaluation)

7,137.8Score (p_act=0.1)

7

Mar 10, 2026

885Point Mass Easy

7

Feb 26, 2026

HalfCheetah-Velocity Safety-Gymnasium (test)

2,468.78Reward

7

Feb 26, 2026

MuJoCo Suite Aggregate

98.7Average Normalized Score

7

Feb 26, 2026

InvertedPendulum v2

1,000Average Return

7

Feb 26, 2026

InvertedDouble v2

9,359Average Return

7

Feb 26, 2026

-4Average Return

7

Feb 26, 2026

4,963Average Return

7

Feb 26, 2026

3,412Average Return

7

Feb 26, 2026

10,352.58Average Return

7

Feb 26, 2026

3,458.22Average Return

7

Feb 26, 2026

98.74Average Return

7

May 22, 2026

DMControl GB (train)

872Cartpole Swingup Score

7

Feb 26, 2026

MuJoCo HalfCheetah 10D-task (c)

1,617Mean Return

7

Feb 26, 2026

MuJoCo HalfCheetah 10D-task (b)

1,984Mean Return

7

Feb 26, 2026

MuJoCo HalfCheetah 10D-task (a)

1,893Mean Return

7

Feb 26, 2026

MuJoCo HalfCheetah Body (test)

1,655Mean Return

7

Feb 26, 2026

MuJoCo HalfCheetah Mass (test)

1,726Mean Return

7

Feb 26, 2026

3,424Hopper Score

7

Feb 26, 2026

InvertedDoublePendulum v1 (train)

9,355.52Max Average Return

7

Feb 26, 2026

InvertedPendulum v1 (train)

1,000Max Average Return

7

Feb 26, 2026

Reacher v1 (train)

-3.6Max Avg Return

7

Feb 26, 2026

4,372.44Max Average Return

7

Feb 26, 2026

Walker2d v1 (train)

4,682.82Max Return

7

Feb 26, 2026

Page 8 of 21

Popular datasets

MuJoCo DeepMind Control Suite Ant MuJoCo HalfCheetah halfcheetah

Follow for updates

Join SOTA2 Contact