An integral part of linking a multi-organ post-absorptive model is the prediction of nutrient fluxes between organs via blood flow. This paper reports a multivariate meta-analysis approach to model portal vein blood flow (PORBF) and hepatic venous blood flow (HEPBF) simultaneously. The developmental database consisted of 296 measurements (pAH dilution) for growing and lactating cattle with 55 treatments from 17 studies, and a separate evaluation database with 31 treatment means from 8 studies. Both databases had information on feed intake, bodyweight and diet composition. Blood flows predicted with DMI or metabolizable energy intake (MEI) was tested with both linear and quadratic equations using the NLINMIX macro of SAS. Cow(study) and study were treated as random effects and blood flow location (PORBF or HEPBF) as a repeated effect. Equations based on DMI rather than MEI typically resulted in higher concordance correlation coefficient (CCC) values, indicating better predictions. Quadratic equations did not out-perform their linear counterparts (CCC analysis), and quadratic equation terms were frequently non-significant. The best predictive equations were: PORBF (L/d) = 4855(±1097) + 2007(±74.8) × DMI (kg/d) and HEPBF (L/d) = 4463(±1094) + 2492(±74.5) × DMI (kg/d), with CCC values of 0.887 and 0.922, respectively. The residuals (predicted – observed) for PORBF expressed as a fraction of HEPBF (PORBF/HEPBF), and hepatic arterial blood flow (ARTBF (L/d); = HEPBF – PORBF) were affected by the proportion of forage in the diet, and thus equations for PORBF and HEPBF based on forage and concentrate DMI were developed: PORBF (L/d) = 5043(±1186) + 1989(±148.8) × Forage (kg DM/d) + 1989(±141.4) × Concentrate (kg DM/d), and HEPBF (L/d) = 4416(±1177) + 2223(±145.5) × Forage (kg DM/d) + 2741(±137.5) × Concentrate (kg DM/d), where CCC values were 0.886 and 0.912, respectively. The CCC for ARTBF improved from 0.877 to 0.904 and PORBF/HEPBF from 0.115 to 0.447 with forage and concentrate DMI separation. Developed equations predicted blood flow well, and also suggest different sensitivity of PORBF and HEPBF to the downstream effects of DMI composition.