The aim of this study was to evaluate the association between cardiorespiratory fitness (CRF) and long-term survival in United States (US) Veterans undergoing CABG. We identified 14,550 US Veterans who underwent CABG at least six months after completing a symptom-limited exercise treadmill test (ETT) with no evidence of cardiovascular disease. During a mean follow-up period of 10.0 ± 5.4 years, 6502 (43.0%) died. To assess the association between CRF and risk of mortality, we formed the following five fitness categories based on peak workload achieved (metabolic equivalents or METs) prior to CABG: Least-Fit (4.3 ± 1.0 METs (n = 4722)), Low-Fit (6.8 ± 0.9 METs (n = 3788)), Moderate-Fit (8.3 ± 1.1 METs (n = 2608)), Fit (10.2 ± 0.8 METs (n = 2613)), and High-Fit (13.0 ± 1.5 METs (n = 819)). Cox proportional hazard models were used to calculate risk across CRF categories. The models were adjusted for age, body mass index, race, cardiovascular disease, percutaneous coronary intervention prior to ETT, cardiovascular medications, and cardiovascular disease risk factors. P-values < 0.05 using two-sided tests were considered statistically significant. The association between cardiorespiratory fitness and mortality was inverse and graded. For every 1-MET increase in exercise capacity, the mortality risk was 11% lower (HR = 0.89; CI: 0.88–0.90; p < 0.001). When compared to the Least-Fit category (referent), mortality risk was 22% lower in Low-Fit individuals (HR = 0.78; CI: 0.73–0.82; p < 0.001), 31% lower in Moderate-Fit individuals (HR = 0.69; CI: 0.64–0.74; p < 0.001), 52% lower in Fit individuals (HR = 0.48; CI: 0.44–0.52; p < 0.001), and 66% lower in High-Fit individuals (HR = 0.34; CI: 0.29–0.40; p < 0.001). Cardiorespiratory fitness is inversely and independently associated with long-term mortality after CABG in Veterans referred for exercise testing.