Abstract Background Albumin-to-neutrophil-lymphocyte ratio (ANLR) is a novel composite biomarker integrating nutritional and inflammatory status. However, its prognostic value for mortality in the general population remains unclear. This study aimed to evaluate the predictive utility of ANLR for all-cause and cardiovascular mortality. Methods This study included 36 628 adults from the National Health and Nutrition Examination Survey 2003–2018. Mortality details were ascertained from the National Death Index. The relationship between ANLR and all-cause and cardiovascular mortality was verified using restricted cubic spline (RCS), weighted Cox proportional hazards model, subgroup analysis and time-dependent receiver operating characteristic curve (ROC). Results RCS analysis revealed an L-shaped ANLR-mortality relationship with an inflection point at 2.19. Below this threshold, each 0.1-unit ANLR increase was associated with 5.0% lower all-cause mortality (HR 0.95, 95% CI 0.94–0.96) and 6.0% lower cardiovascular mortality (HR 0.94, 95% CI 0.92–0.96). Participants were stratified into higher (> 1.23) and lower (≤ 1.23) ANLR groups. Weighted Cox proportional hazards models demonstrated that individuals with higher ANLR had a significantly reduced risk of all-cause (HR 0.57, 95% CI 0.53–0.62) and cardiovascular mortality (HR 0.55, 95% CI 0.47–0.65). Results were consistent across subgroups. Time-dependent ROC analysis confirmed moderate predictive ability over 1–7 years, with area under the curve values of 0.667–0.708 for all-cause and 0.690–0.703 for cardiovascular mortality. Conclusion Elevated ANLR is associated with reduced mortality risk, though non-linearly. Clinical attention to albumin supplementation and maintaining appropriate neutrophil-to-lymphocyte ratio levels may be warranted in high-risk populations.