With Motor-Imagery (MI) Brain--Machine Interface (BMI) we may control machines by merely thinking of performing a motor action. Practical use cases require a wearable solution where the classification of the brain signals is done locally near the sensor using machine learning models embedded on energy-efficient microcontroller units (MCU), for assured privacy, user comfort, and long-term usage. In this work, we provide practical insights on the accuracy-cost trade-off for embedded BMI solutions. Our proposed Multispectral Riemannian Classifier (MRC) reaches 75.1% accuracy on 4-class MI task. We further scale down the model by quantizing it to mixed-precision representations with a minimal accuracy loss of 1%, which is still 3.2% more accurate than the state-of-the-art embedded Convolutional Neural Network. We implement the model on a low-power MCU with parallel processing units taking only 33.39ms and consuming 1.304mJ per classification.