Background Robotic gastrectomy (RG) for gastric cancer (GC) has been increasingly performed for a decade; however, evidence for its use as a standard treatment has not yet been established. The present study aimed to determine the safety, feasibility, and effectiveness of RG for GC. Methods This multi-institutional, single-arm prospective study, which included 330 patients from 15 institutions, was designed to compare morbidity rate of RG with that of a historical control (conventional laparoscopic gastrectomy, LG). This trial was approved for Advanced Medical Technology (“Senshiniryo”) B. The included patients were operable patients with cStage I/II GC. The primary endpoint was morbidity (Clavien–Dindo Grade ≥ IIIa). The specific hypothesis was that RG could reduce the morbidity rate to less than half of that with LG (6.4%). A sample size of 330 was considered sufficient (one-sided alpha 0.05, power 80%). Results Among the 330 study patients, the protocol treatment was suspended in 4 patients. Thus, 326 patients fully enrolled and completed the study. The median patient age and BMI were 66 years and 22.4 kg/m 2 , respectively. Distal gastrectomy was performed in 253 (77.6%) patients. The median operative time and estimated blood loss were 313 min and 20 mL, respectively. No 30-day mortality was seen, and morbidity showed a significant reduction to 2.45% with RG ( p  = 0.0018). Conclusions RG for cStage I/II GC is safe and feasible. It may be effective in reducing morbidity with LG.