Objective The safety of weight loss and low weight gain during pregnancy remains unclear. To determine how different patterns of gestational weight gain (GWG), including weight loss, stability, and low GWG relate to perinatal outcomes by prepregnancy obesity class. Study Design The study population included 29,408 singleton livebirths among pregnant people with obesity from Kaiser Permanente Northern California (2008–2013). Clinically measured GWG was grouped into meaningful categories (Adequate: reference, met 2009 National Academy of Medicine [NAM] Guidelines [5–9.1 kg], Excessive [>9.1 kg], Low [1–4.9 kg], Stable [±1 kg], Weight Loss [>1 kg]) or GWG Z-score quintiles. Modified Poisson regression was used to estimate risk of adverse outcomes, stratified by obesity class. Electronic health record data were used to define outcomes, including cesarean delivery, preterm birth, admission to the neonatal intensive care unit, small- and large-for-gestational age infants. Results Prevalence of weight stability and weight loss was 3.8 and 3.4%, respectively. Compared with those who gained within NAM, increased risk of small-for-gestational age was observed among those with weight loss among obesity class I (Risk Ratio (RR): 1.57, 95% confidence interval [CI]: 1.12, 2.19), obesity class II (RR: 2.18, 95% CI: 1.52, 3.13), and obesity class III (RR: 1.72, 95% CI: 1.21, 2.45). Weight loss was associated with a decreased risk of cesarean delivery among obesity class III, compared with NAM. Conclusion Weight loss during pregnancy is associated with increased risk of small-for-gestational age among all obesity classes, but not other adverse perinatal outcomes and may reduce risk of cesarean delivery. Low weight gain and weight stability are not associated with risk of adverse outcomes among those with class III obesity. GWG guidelines may need to vary by obesity class. Key Points