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

DeepMind Control dataset and benchmarks · SOTA2 Research

Research Datasets

DeepMind Control

Benchmarks

Task NameDataset NameSOTA MethodSort

Task NameDataset NameSOTA ResultTrendResults

Reinforcement Learning

DeepMind Control Cartpole Balance Sparse

150,700Steps to 75% Return

11

Behavior Cloning

DeepMind Control (DMC) suite seen/unseen embodiments

49.1Hopper Hop Score

9

Reinforcement Learning

DeepMind Control Reacher Hard

589.7Steps to 75% Return (k)

8

Vision-based robot policy learning

DeepMind Control

89.1Stand Performance

8

Reinforcement Learning

DeepMind Control Walker Walk

116.8Steps to 75% Return

6

Reinforcement Learning

DeepMind Control Walker Stand

120Steps to 75% Return

6

Reinforcement Learning

DeepMind Control Reacher Easy

205,000Steps to 75% Return Threshold

6

Reinforcement Learning

DeepMind Control Quadruped Walk

330,500Steps to Reach 75% Return

6

Reinforcement Learning

DeepMind Control Pendulum Swingup

76,600Steps to 75% Return (k)

6

Reinforcement Learning

DeepMind Control Hopper Hop

153,500Steps to 75% Return

6

Reinforcement Learning

DeepMind Control Finger Turn Hard

333.4Steps to 75% Return

6

Reinforcement Learning

DeepMind Control Finger Turn Easy

77,300Steps to 75% Return

6

Reinforcement Learning

DeepMind Control Finger Spin

89,100Steps to 75% Return

6

Reinforcement Learning

DeepMind Control Cup Catch

104,700Steps to 75% Return

6

Reinforcement Learning

DeepMind Control Cartpole Swingup

278,800Steps to 75% Return

6

Reinforcement Learning

DeepMind Control Cartpole Balance

203,100Steps to 75% Return

6

Reinforcement Learning

DeepMind Control Acrobot Swingup

475,000Steps to 75% Return

6

Reinforcement Learning

DeepMind Control Walker Run

205,000Steps to 75% Return

5

Reinforcement Learning

DeepMind Control Hopper Stand

245.2Steps to 75% Return (k)

5

Reinforcement Learning

DeepMind Control Cheetah Run

145,400Environment Steps to 75% Return (k)

5

Reinforcement Learning

DeepMind Control Cartpole Swingup Sparse

383,100Environment Steps to 75% Return

5

Continuous Control

DeepMind Control 3M steps

422Acrobot Swingup Score

5

Continuous Control

DeepMind Control 1.5M steps

272Acrobot Swingup Score

5

Reinforcement Learning

DeepMind Control Quadruped Run

345,000Steps to 75% Return

4

Continuous Control

DeepMind Control

89.1Walker-stand Success Rate

4

Page 1 of 2

PreviousNext

Popular tasks

Reinforcement Learning Continuous Control Control Behavior Cloning Vision-based robot policy learning

Follow for updates

Join SOTA2 Contact