Current gestational weight change (GWC) recommendations for obese individuals were established with limited evidence of the pattern and timing of weight change across pregnancy. Similarly, the recommendation of 5-9 kg does not differentiate by the severity of obesity. We sought to describe GWC trajectory classes by obesity grade and associated infant outcomes among a large, diverse cohort. The study population included 22,355 individuals with singleton pregnancies, obesity (BMI ≥30.0 kg/m2), and normal glucose tolerance who delivered at Kaiser Permanente Northern California between 2008 and 2013. Obesity grade-specific GWC trajectories were modeled at 38 wk using flexible latent class mixed modeling (package lcmm) in R. Multivariable Poisson or linear regression models estimated the associations between the GWC trajectory class and infant outcomes (size-for-gestational age and preterm birth) by obesity grade. Five GWC trajectory classes were identified for each obesity grade, each with a distinct pattern of weight change before 15 wk (including loss, stability, and gain) followed by weight gain thereafter (low, moderate, and high). Two classes with high overall gain were associated with an increased risk for large for gestational age (LGA) in obesity grade 1 (IRR = 1.27; 95% CI: 1.10, 1.46; IRR = 1.47; 95% CI: 1.24, 1.74). Both high (IRR = 2.02; 95% CI: 1.61, 2.52; IRR = 1.98; 95% CI: 1.52, 2.58) and 2 moderate-gain classes (IRR = 1.40; 95% CI 1.14, 1.71; IRR = 1.51; 95% CI: 1.20, 1.90) were associated with LGA in grade 2, and only early loss/late moderate-gain class 3 (IRR = 1.30; 95% CI: 1.04, 1.62) was associated in grade 3. This class was also associated with preterm birth in grade 2. No associations were detected between GWC and small for gestational age (SGA). Among the pregnancies affected by obesity, GWC was not linear or uniform. Different patterns of high gain were associated with an increased risk for LGA with the greatest magnitude in obesity grade 2, whereas GWC patterns were not associated with SGA.