Recent contributions to the financial econometrics literature exploit high-frequency (HF) data to improve models for daily asset returns. This paper proposes a new class of dynamic extreme value models that profit from HF data when estimating the tails of daily asset returns. Our Realized Peaks Over Threshold approach provides estimates for the tails of the time-varying conditional return distribution, and remains applicable even when data are non-stationary. An in-sample fit to S&P 500 index returns shows that HF data convey information on daily extreme returns beyond that included in low frequency data. Finally, our out-of-sample forecasts of conditional risk measures show good performance.