Hybrid equids such as mules and hinnies play a critical role in supporting livelihoods in low- and middle-income countries but remain underrepresented in veterinary research. One key challenge is the absence of a practical and internally validated method for estimating body weight in mules and hinnies, which is essential for accurate dosing and management, particularly in the absence of a weighbridge. This study aimed to evaluate the performance of existing equid weight estimation models on mules and hinnies and to develop a new mules and hinnies -specific model to improve predictive accuracy and field usability. Three hundred and sixteen healthy mules and hinnies were assessed across nine countries. Sixteen morphological measurements were collected, and body weight was recorded using a digital weighbridge. Existing models were tested for their feasibility, and linear regression model was developed to predict mules and hinnies’ weighbridge body weight from morphological measurements. Model performance was assessed using repeated cross-validation and standard error metrics. Machine learning methods were used as a secondary analysis to explore feature importance. Morphological predictors showed strong associations with weighbridge weight. Linear models demonstrated reliable predictive performance, with low prediction error across weight ranges. The new mules and hinnies-specific model based on heart girth, chest width, and forearm girth, achieved an adjusted r² of 0.94 (p < 0.001), providing estimates of live-weight without requiring classification by sex, conformation, or type. These findings demonstrate that simple morphological measurements can provide good estimates of body weight of mules and hinnies in field settings, greatly supporting the daily management of mules and hinnies in different contexts, and contribute to improve their health and welfare globally.