Air pollution is a global problem and has a severe impact on human health. Fine-grained air quality (AQ) monitoring is important in mitigating air pollution. However, existing AQ station deployments are sparse due to installation and operational costs. In this work, we propose an Active Learning-based method for air quality station deployment. We use Gaussian processes and several classical machine learning algorithms (Random Forest, K-Neighbors and Support Vector Machine) to benchmark several active learning strategies for two real-world air quality datasets (Delhi, India and Beijing, China).