BACKGROUND: Metabolic effects of omega-6 polyunsaturated fatty acids (n-6 PUFA) remain contentious, and evidence is limited regarding their potential role in primary prevention of type 2 diabetes (T2D). METHODS: Global consortium of 20 prospective cohort studies with new, harmonized, individual-level analyses. Each study measured biomarker linoleic acid (LA, the major dietary PUFA) and arachidonic acid (AA, a precursor of metabolic- and inflammation-mediating metabolites) at baseline among adults without prevalent T2D. Associations of n-6 PUFA biomarkers with T2D risk were assessed prospectively using a pre-specified analytic plan for exposures, covariates, and effect modifiers. Findings were pooled using inverse-variance weighted meta-analysis. FINDINGS: This study comprised 39,740 adults from 10 countries, with age (range of cohort means) 49-76yrs and BMI, 23∙3-28∙4kg/m2, including 4,347 incident T2D cases during 366,073 person-years of follow-up. In multivariable-adjusted pooled analyses, higher LA biomarker associated with 35% lower risk of T2D (RR per interquintile range: 0∙65, 95%CI: 0∙60-0∙72, P<0∙001, I2=54%). Findings for LA biomarker were generally similar in different lipid compartments including phospholipids, plasma, cholesterol esters, and adipose tissue. Levels of AA biomarker were not significantly associated with T2D risk overall (RR per interquintile range: 0∙96, 95%CI: 0∙88-1∙05, P=0∙38, I2=63%). Relations of LA and AA biomarker with T2D were not significantly modified by age, BMI, sex, race, aspirin use, n-3 PUFA levels, or FADS genetic variants (all P-interaction≥0∙13). INTERPRETATION: Findings suggest long-term benefit of LA, and no harms of AA, for the prevention of T2D.