Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Visual Reinforcement Learning on DeepMind Control Suite 500K steps
Loading...
408
Quadruped Walk Score
ResWM
271.76
307.13
342.5
377.87
Mar 11, 2026
Quadruped Walk Score
Pendulum Swingup Score
Hopper Hop Score
Finger Turn Hard Score
Walker Run Score
Average Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
Quadruped Walk Score
Pendulum Swingup Score
Hopper Hop Score
Finger Turn Hard Score
Walker Run Score
Average Score
ResWM
Environment steps=500K...
2026.03
408
645
116
513
496
435.6
ResAct
Environment steps=500K...
2026.03
385
618
99
465
467
406.8
TACO
Environment steps=500K...
2026.03
345
485
112
372
355
333.8
DeepRAD
Environment steps=500K...
2026.03
307
308
51
173
375
242.8
Flare
Environment steps=500K...
2026.03
296
242
90
282
426
267.2
MaDi
Environment steps=500K...
2026.03
277
372
80
311
382
284.4
Feedback
Search any
task
Search any
task
