Objective To explore whether the effect of cloud rehabilitation system on motor dysfunction and activities of daily living (ADL) of stroke patients is not inferior to hemiplegia manual treatment. Methods This study adopted a multicenter randomized controlled trial design. A total of 118 stroke patients were enrolled from 5 hospitals in China between April 2018 and April 2019, and they were randomly divided into a trial group and a control group, with 59 patients in each group. The trial group adopted hemiplegia manual treatment (≥30 min per time, once a day) and cloud rehabilitation training (30 min per time, once a day), and the control group adopted hemiplegia manual treatment alone (≥30 min per time, twice a day). All patients received continuous treatment for 2 weeks, and followed up for 2 weeks after that. The Brunnstrom stage (BRS), IK exercise stage, and Modified Barthel Index (MBI) were used to evaluate the motor function and ADL before and 4 weeks after treatment. Results There was no significant difference in gender, age, course, hemiplegia side, source (inpatient/outpatient) or hospital grade between the two groups (P>0.05), which were comparable. Compared with those before treatment, there were significant improvements in the BRS (upper and lower extremities), IK exercise stages (upper and lower extremities), and MBI scores in both groups 4 weeks after treatment (P<0.05). The four-week improvements in motor function indicators (BRS and IK) of the trial group were not inferior to those of the control group (P<0.05), and there was no significant difference in MBI score between the two groups 4 weeks after treatment (F=1.498, P>0.05). Conclusion The cloud rehabilitation system is not inferior to hemiplegia manual treatment in improving the limb motor function or ADL of patients .