Prediction models for enteric methane (CH4) emissions from beef cattle proposed by various groups may not perform with similar accuracy for the low- and middle-income countries in South-east Asia (SE-Asia) because beef cattle in these countries are raised under different climatic conditions with diverse feeding systems, and have different CH4 emission characteristics. The objectives of this study were to: i) predict CH4 emission (g d−1 animal−1), yield [g kg−1 dry matter intake; DMI)−1], intensity [g kg−1 average daily gain)−1], and CH4 conversion factor (Ym) using an intercountry database of individual animal records from SE-Asia; ii) evaluate the impact of different dietary forage contents (all-, high- and low-forage) representing the diverse feeding systems on CH4 emission, yield, intensity and Ym in SE-Asia; and iii) cross-validate equations from this study with published data. A total of 398 individual animal observations of beef cattle from SE-Asia were used for this analysis. Linear models developed by incrementally adding covariates revealed that CH4 emission model using only DMI fitted to all data had a root mean square prediction error (RMSPE) of 16.9%. Subsets containing data with 100% forage in the diet (all-forage), 50–85% (high-forage) and < 50% (low-forage) had an RMSPE of 16.5%, 14.7%, and 17.4%, respectively. Linear multiple equation based on DMI and dietary NDF concentration (DMI + NDF_C, RMSPE = 15.2%; all-data) improved prediction accuracy over that of DMI alone. The DMI + NDF_C models for all-forage (RMSPE = 14.6%) and high-forage subsets (RMSPE = 13.3%) except for low-forage (RMSPE = 16.4%), improved the precision and accuracy of CH4 emission prediction. Methane yield and CH4 emission intensity could not be reliably modelled with the current database. The present study provides improved CH4 prediction models for beef cattle managed under diverse feeding systems in SE-Asia and affirmed that region-specific models are needed to reliably predict beef cattle CH4 emission at national or regional levels, particularly for low- and middle-income countries.