CEBRA, or Learnable Latent Embeddings for Joint Behavioural and Neural Analysis, is a cutting-edge machine-learning tool designed to map behavioral actions to neural activity, a crucial objective in neuroscience research. This innovative method stands out for its ability to utilize both behavioral and neural data to create high-performance latent spaces, revealing underlying behavior correlates. CEBRA is versatile, capable of analyzing single and multi-session datasets, handling calcium and electrophysiology data, decoding natural movies from the visual cortex, and reconstructing viewed videos from the mouse brain. With features like space mapping, kinematic feature analysis, and rapid decoding, CEBRA is a valuable asset for neuroscience and behavioral studies.