Home/Browse/Reinforcement Learning/Continuous Control/MuJoCo

MuJoCo

Name: MuJoCo Benchmark Results
Creator: Google DeepMind
License: https://creativecommons.org/licenses/by/4.0/

Google DeepMind

Physics engine for continuous control tasks like walking, running, and manipulation.

Paper Leaderboard

Benchmark Stats

Models12

Papers12

Metrics1

SOTA History

Not enough data to show trend.

Average Return

Mean episodic return averaged across HalfCheetah, Hopper, and Walker2d at 1M steps.

Higher is better

Rank	Model	Source	Score	Year	Paper
1	TD3 Mean of HalfCheetah-v4 (9583), Hopper-v4 (3134), Walker2d-v4 (4057) at 1M steps. CleanRL verified.	Community	5592	2026	Source
2	SAC Mean of HalfCheetah-v4 (9634), Hopper-v4 (2310), Walker2d-v4 (3591) at 1M steps. CleanRL verified.	Community	5179	2026	Source
3	PPO Mean of HalfCheetah-v4 (1442), Hopper-v4 (2382), Walker2d-v4 (2287) at 1M steps. CleanRL verified.	Community	2038	2026	Source
4	TD-MPC2 (317M params) TD-MPC2, 317M-param shared model. Mean normalized score across 15 DMControl tasks, 1M steps. ICLR 2024.	Editorial	960	2026	Source
5	TD-MPC2 (19M params) TD-MPC2, 19M-param shared model. Mean normalized score across 15 DMControl tasks, 1M steps. ICLR 2024.	Editorial	953	2026	Source
6	FOWM FOWM (Foundation Online World Models). Mean normalized score, DMControl 15 tasks. NeurIPS 2024.	Editorial	945	2026	Source
7	BRO BRO (Best-of-N Robustness RL). Mean normalized score across DMControl tasks. ICML 2024.	Editorial	941	2026	Source
8	TD-MPC2 (5M params) TD-MPC2, 5M-param model. Mean normalized score across 15 DMControl tasks, 1M steps. ICLR 2024.	Editorial	929	2026	Source
9	DreamerV3 DreamerV3. Mean normalized score across 15 DMControl tasks, 1M steps. From TD-MPC2 Table 1 comparison.	Editorial	897	2026	Source
10	TD-MPC TD-MPC (original). Mean normalized score across DMControl tasks, 1M steps. ICML 2022 baseline from TD-MPC2 paper.	Editorial	857	2026	Source
11	DrQ-v2 DrQ-v2, pixel-based. Mean normalized score across 15 DMControl tasks, 1M steps. From TD-MPC2 Table 1.	Editorial	799	2026	Source
12	SAC (state-based) SAC (state-based). Mean normalized score across DMControl tasks. Classic baseline from TD-MPC2 Table 1.	Editorial	777	2026	Source

Submit a Result

Back to Continuous Control