Objective We examined the dose‐response relationship between weight reduction and pain/functional improvement in persons with symptomatic knee osteoarthritis (KOA) participating in a community‐based weight loss program. Methods Consecutive participants with KOA and enrolled in the 18‐week Osteoarthritis Healthy Weight for Life weight‐loss program were selected. In this completer‐type analysis, participants were assessed at baseline, 6 weeks, and 18 weeks for body weight and Knee Injury and Osteoarthritis Outcome Score (KOOS) subscales. The dose‐response relationship between weight‐change categories (>10%, 7.6–10%, 5.1–7.5%, 2.6–5.0%, and <2.5% of body weight loss) and change in KOOS scores was assessed by repeated‐measures analysis of variance, controlling for sex and age, body mass index (BMI), and KOOS. The Western Ontario McMaster Universities Osteoarthritis Index function score derived from the KOOS was used to assess a meaningful clinical functional improvement. Results A total of 1,383 persons (71% females) were enrolled. Mean ± SD age, height, and weight were 64 ± 8.7 years, 1.66 ± 0.09 meters, and 95.1 ± 17.2 kg, respectively. Mean ± SD BMI was 34.4 ± 5.2 kg/m2 with 82% of participants obese at baseline. A total of 1,304 persons (94%) achieved a >2.5% reduction in body weight. There was a significant dose‐response relationship between all KOOS subscales and percentage of weight change across all weight‐change categories. Participants required ≥7.7% (95% confidence interval 5.2, 13.3) body weight loss to achieve a minimal clinically important improvement in function. Conclusion There is a significant dose‐response relationship between percentage of weight loss and symptomatic improvement. This study confirms the feasibility of weight loss as a therapeutic intervention in KOA in a community‐based setting.