Forecasting a price is an essential process to earn more profit in a variety of financial markets, and many machine learning models have been developed but faced problems when conducted: manually stack or manually tune for each different market. However, it is not an easy task because they are intrinsically non-linear and non-stationary systems. An idea that's been around since the 1990s, AutoML has been described as a "quiet revolution in AI" is the process of automating the time-consuming, iterative tasks of ML model development, enable domain experts to automatically build ML applications without much requirement for statistical and ML knowledge. In this work, we propose a novel auto machine learning(AutoML)-based approach to predict multiple types of financial markets. Specifically, it employed 11 learning models, and was tested to predict four types of financial markets in 16 countries. We demonstrated that our method was consistently more accurate and profitable than the baseline and the buy-and-hold strategy, respectively. Taken together, our AutoML approach combined with multi-market information, is useful to predict the financial markets.