Offline Reinforcement Learning on D4RL (MuJoCo M/MR/ME & Adroit/Kitchen Subset) benchmark leaderboard · SOTA2 Research

Offline Reinforcement Learning on D4RL (MuJoCo M/MR/ME & Adroit/Kitchen Subset)

113.3Walker2d (Medium Expert) ScoreICQL

Evaluation Results

Method	Links
ICQL2025.09		113.3	80.3	81.9	111	62.6	96.4	89.1	45.9	44.7	85.6	89.4	3.7	4.5	17.1	11.7	79.3	59.5	61.5	69.7
IQL2025.09		109.8	71.5	61	98.5	63.3	82.4	83.4	42.5	38.9	89.5	84.9	7.2	0.5	9.8	7.6	59.2	53.3	45.8	63.2
DMG2025.09		109.5	85	81.9	109.8	92.3	100.1	93.6	47.9	44.6	66.2	67.5	18.4	13.4	0.1	3.7	22.5	30	37.5	62
TD3BC2025.09		109.2	77	41.5	78.2	53.5	59.4	62.8	43.1	41.8	64.6	76.8	1.5	1.8	0.2	-0.1	57.5	53.5	46.7	47.7
ReBRAC2025.09		109.2	82.8	39.4	98.7	60.6	87.4	84.6	44.6	36.9	91.5	68.9	1.1	0.2	-0.1	9	60	47.5	62.5	60
RADT2025.09		107.8	68.9	67.2	109.4	62.4	81.6	90.9	42	38.9	17.8	32.4	0.7	1.3	13.2	2.4	32.5	54.1	53.8	57.2
BC2025.09		107.5	75.3	26	52.5	52.9	18.1	55.2	42.6	36.6	63.9	37	1.2	0.6	2	0	65	51.5	38	47.3
QC2025.09		102.8	34.1	46.6	44	64.6	18.6	94.2	48.2	40.5	55.7	54.8	1.2	2.2	0.7	4.4	27.5	60	52.5	47.7
FQL2025.09		101	72.4	60.9	60.1	55.6	55	92.9	43.9	40	61.2	23.5	1.1	1.7	0.2	0.1	16.3	45	15.8	50.5
CQL2025.09		98.7	79.2	77.2	105.4	58	95	62.4	44.4	45.5	37.5	39.2	4.4	2.1	9.9	0.1	43.8	51	49.8	58.5
DT2025.09		70.7	70.2	54.8	57.5	57.1	65.8	70.8	42.8	39.5	79.5	74	1.7	3.7	5.5	3.2	52.5	60	55	56.2