Robotic Manipulation on RLBench standard v1 benchmark leaderboard · SOTA2 Research

Robotic Manipulation on RLBench standard v1

81Average Success RateCortical Policy

Evaluation Results

Method	Links
Cortical Policy2026.03		81	1.8	96	100	38.7	100	84	24	94.7	100	65.3	100	98.7	81.3	86.7	37.3	81.3	76	100	94.7
RVT-22026.03		77.5	3.5	93.3	97.3	28	100	92	32	84	100	44	98.7	92	86.7	74.7	26.7	80	72	98.7	94.7
VIHE2026.03		77	3.6	48	100	84	100	76	12	88	100	60	96	92	92	96	52	68	68	64	92
SAM-E2026.03		70.6	3.8	82.4	100	18.4	95.2	95.2	0	94.4	100	64	92	95.2	78.4	95.2	34.4	26.4	0	100	100
Σ-agent2026.03		68.8	4.2	78.4	100	15.2	97.6	76.8	0.8	90.4	100	66.4	70.4	98.4	73.2	74.4	36	51.2	33.6	80.8	95.2
3D-MVP2026.03		67.5	4.3	76	100	20	96	84	4	100	96	60	100	92	60	48	28	40	36	80	96
RVT2026.03		62.9	5.7	52	99.2	11.2	88	71.2	4	91	100	49.6	88	91.2	48	81.6	36	28.8	26.4	72	93.6
PerAct2026.03		49.4	7.6	55.2	89.6	5.6	70.4	88	2.4	44.8	92.8	28	51.2	84	17.6	74	16.8	26.4	2.4	52	88
Hiveformer2026.03		45.3	8.1	52	76	0	100	52	0	80	84	32	68	76	8	64	8	8	0	28	80