AIMS: Sharp-wave ripples (SWR) are high frequency events recorded in the local field potential (LFP) of the hippocampus of rodents and humans. During neurological conditions such as epilepsy and Alzheimer’s disease, changes on the waveforms of SWRs are considered as biomarkers of dysfunction. However, SWR features cannot be fully characterized by spectral methods alone.METHODS: Here, we introduce a machine learning (ML) toolbox resulting from a community-based hackathon, that operates over high-density LFP recordings to detect hippocampal SWR. We train a wide variety of ML architectures (support-vector machines, decision trees, and recurrent and convolutional neural networks) using data from the dorsal hippocampus of mice, and extend detection to macaque data.RESULTS: We show how each method optimises detection using different time windows and number of channels, but all perform better when spatial information is provided. We report that 1D-convolutional and recurrent neural networks have better performance and are more robust than the other ML architectures. When allowing them to work together in an ensemble, optimal performance is achieved. Finally, we successfully detect SWRs without re-training using non-human primates, revealing common LFP features across species.CONCLUSIONS: We have developed an open source toolbox with several ML architectures to detect hippocampal SWRs. In addition to detection, this toolbox offers the possibility of retraining all ML architectures to fit new data, easing its usage for other species, different conditions, or other HFOs events. We conclude this approach can be used as a discovery tool for better understanding the dynamics of SWR.