The availability of rapid and reliable tools for monitoring of plants' cold tolerance is a prerequisite for research aimed at breeding of cold-tolerant crop plants. Therefore, we have tested the capacity of metabolomics-based methods employing ultra-high-performance liquid chromatography (UHPLC)-mass spectrometry and direct analysis in real time-mass spectrometry for high-throughput screening of cold tolerance in eight differentially cold-tolerant accessions of Arabidopsis thaliana. Metabolomic fingerprinting of leaf tissues was performed in methanolic extracts for (1) 6-week-old non-acclimated (NAC) plants grown at room temperature, (2) NAC plants cold-acclimated (ACC) at 4 °C for 2 weeks, and (3) cold-acclimated plants given sub-zero-temperature treatments by slow cooling at -4 °C for 8 h. The generated chromatograms and mass spectra were processed with the use of multivariate statistical analysis employing principal component analysis (PCA) and linear discriminant analysis. The PCA of metabolomic fingerprints classified the investigated A. thaliana accessions into three categories with low, intermediate, and high cold tolerance for both the cold-acclimated and the sub-zero-temperature-treated plants. This indicates the potential application of metabolomics-based fingerprinting for measuring cold tolerance in the cold-acclimated state, i.e., without treating plants at freezing temperatures that is required by currently available methods. Furthermore, we employed UHPLC coupled to the quadrupole-time-of-flight mass spectrometry to identify characteristic metabolites in ACC state and found the abundance of gluconapin and flavon-3-ol glycosides, respectively, in the cold-sensitive and the cold-tolerant accessions.