Offline Reinforcement Learning on D4RL MuJoCo (Medium, Med-Replay, Medium-Expert Configurations) benchmark leaderboard · SOTA2 Research

Offline Reinforcement Learning on D4RL MuJoCo (Medium, Med-Replay, Medium-Expert Configurations)

71.9HalfCheetah (m)CAC-2

Evaluation Results

Method	Links
CAC-22026.06		71.9	99.7	84.1	62.7	100.4	83	89.2	106	111.6	89.8
MPCwDWM2026.03		70.05	103.38	88.91	59.89	103.14	95.87	105.35	107.42	115.76	94.4
BFQ2026.06		66.1	103.5	91.7	52.1	102.1	97.5	98.6	110.5	113.4	92.8
FAC2026.02		65	91.9	85.2	55.4	99.1	83	101.9	104.2	108.4	88.2
OFQL2026.06		63.8	103.6	87.4	51.2	101.9	106.2	95.2	110.2	113	92.5
CAC2026.06		62.2	46.3	86.6	47.9	91.7	86.6	47.9	0.1	-1.3	52
SRPO2026.06		60.4	95.5	84.4	51.4	101.2	84.6	92.2	100.1	114	87.1
FQL2026.06		60.1	74.5	72.7	51.1	85.4	82.1	99.8	86.2	100.5	79.2
SORL2026.06		57.4	81.3	77.9	48.3	93	67.8	96.5	45.9	109.1	80
DQL2026.06		52.3	96.5	86.8	47.9	91.2	98.2	95.5	111.1	111.6	89
IDQL2026.06		51.5	70.1	86.1	46.5	99.4	89.1	91.3	110.1	110.6	84.1
EDP2026.06		50.8	72.6	86.5	44.9	83	87	95.8	110.8	110.4	82.4
TD3-BC2026.06		48.5	56.6	83.3	44.5	55.2	82.5	91.5	101.6	110.4	74.9
IQL2026.06		47.7	61.2	78.7	42.9	86.8	68.3	88.3	76.6	108.7	73.2
TT2026.03		46.9	61.1	79	41.9	91.5	82.6	95	110	101.9	78.9
MBOP2026.03		44.6	48.8	41	42.3	12.4	9.7	105.9	55.1	70.2	47.8
DM2026.02		43.8	98.5	80.3	40.8	89.1	79.3	93.5	111.9	111.6	83.2
DC2026.02		43	92.5	79.2	41.3	94.2	76.6	93	110.4	109.6	82.2
Reinformer2026.02		42.9	81.6	80.5	39	83.3	72.9	92	107.8	109.4	78.8
Diffuser2026.03		42.8	74.3	79.6	37.7	93.6	70.6	88.9	103.3	106.9	77.5
DT2026.02		42.6	67.6	74	36.6	82.7	66.6	86.8	107.6	108.1	74.7
DT2026.03		42.6	67.6	74	36.6	82.7	66.6	86.8	107.6	108.1	74.7
BC2026.06		42.5	52.9	72.3	34.7	19.7	29	60.2	67.2	105.4	53.8
QDT2026.02		42.3	66.5	67.1	35.6	52.1	58.2	—	—	—	—
MOPO2026.03		42.3	28	17.8	53.1	67.5	39	63.3	23.7	44.6	42.1
MOReL2026.03		42.1	95.4	77.8	40.2	93.6	49.8	53.3	108.7	95.6	72.9