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

Reinforcement Learning research benchmarks – Page 19 · SOTA2 Research

Reinforcement Learning

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

Bullet Safety Gym AntRun

0.65Reward

6

Feb 26, 2026

Bullet Safety Gym CarRun

0.98Reward

6

Feb 26, 2026

Bullet Safety Gym BallRun

0.59Reward

6

Feb 26, 2026

Safety Gym AntVel

1Reward

6

Feb 26, 2026

Safety Gym Walker2dVel

0.79Reward

6

Feb 26, 2026

Safety Gym HalfCheetahVel

0.97Reward

6

Feb 26, 2026

Safety Gym HopperVel

0.7Reward

6

Feb 26, 2026

Safety Gym SwimmerVel

0.63Reward

6

Feb 26, 2026

BipedalWalker v3

273.2Return

6

Mar 12, 2026

Office World Map 3

25,222Steps to 100% Success

6

Feb 26, 2026

Office World Map 2

16,240Training Steps

6

Feb 26, 2026

Office World Map 1

7,312Training Steps

6

Feb 26, 2026

DMC-GB2 Color Video Hard (test)

437Cartpole Swingup Score

6

Feb 26, 2026

Recommender discrete

DGRL (SDN, DBU)

2,009.88Mean Score

6

Feb 26, 2026

Atari Human Normalized 100M frames

QRC(λ) + SPR + Orthogonal Gradient Updates

1.61Mean Return

6

Feb 26, 2026

Taxi environment stochastic

-57Episode Return

6

Feb 26, 2026

24Mean Score

6

Feb 26, 2026

22.7Mean Performance

6

Feb 26, 2026

85.1Mean Performance

6

Feb 26, 2026

34.1Mean Score

6

Feb 26, 2026

32.9Mean Performance Score

6

Feb 26, 2026

1,000Average Decisions

6

Feb 26, 2026

D4RL AntMaze v0 (online fine-tuning)

96.9Success Rate (umaze-v0)

6

Feb 26, 2026

D4RL Locomotion online fine-tuning v2

97.6Hopper v2 MR Score

6

Feb 26, 2026

35Overall Score

6

Feb 26, 2026

Page 19 of 41

Popular datasets

Atari Atari 2600 halfcheetah DMControl hopper

Follow for updates

Join SOTA2 Contact