ω‐3 polyunsaturated fatty acids (PUFAs) and ω‐6 PUFAs have opposing influences on inflammation. The objective was to determine whether lipopolysaccharide (LPS)–induced cytokine release by human alveolar cells was affected by changes in the ω‐3/ω‐6 ratio of cell membranes induced by different supplies of PUFAs. After LPS challenge, PUFAs were added to alveolar cells as docosahexaenoic acid (DHA, ω‐3) and arachidonic acid (AA, ω‐6) in 4 different DHA/AA ratios (1:1, 1:2, 1:4, and 1:7), and the effects on cytokine release were measured. The supply of 1:1 and 1:2 DHA/AA ratios reversed the baseline predominance of ω‐6 over ω‐3 in the ω‐3/ω‐6 PUFA ratio of cell membranes. The release of proinflammatory cytokines (tumor necrosis factor α, interleukin‐6, and interleukin‐8) was reduced by 1:1 and 1:2 DHA/AA ratios (P < .01 to P < .001) but increased by 1:4 and 1:7 DHA/AA ratios (P < .01 to P < .001) vs control. The 1:1 and 1:2 ratios increased the release of anti‐inflammatory interleukin‐10 (P < .001). The balance between proinflammatory and anti‐inflammatory cytokines showed an anti‐inflammatory response with 1:1 and 1:2 ratios and a proinflammatory response with 1:4 and 1:7 ratios (P < .001). This study showed that proinflammatory cytokine release was dependent on the proportion of ω‐3 in the ω‐3/ω‐6 ratio of alveolar cell membranes, being reduced with the supply of a high proportion of DHA and increased with a high proportion of AA, respectively. These results support the biochemical basis for current recommendations to shift the PUFA supply from ω‐6 to ω‐3 in nutrition support of patients with acute lung injury.