In this study, metabolic profiles of a set of 48 rice germplasms from the Chinese core collection were obtained by gas chromatography and time-of-flight mass spectrometry (GC-TOF-MS). Forty-one metabolites were identified and relatively quantified according to the internal standard (IS). Wide ranges of variations for all metabolites were observed among rice accessions. The maximum/minimum ratios varied from 4.73 to 211.36. The metabolites were categorized into seven groups based on their chemical characteristics. Clustering analysis and a correlation network showed that most of the metabolites had variations among rice accessions in the same direction. Using 218 molecular markers, association mapping was conducted to identify the chromosomal loci influencing the concentrations of identified metabolites. Twenty markers were identified associating with the concentrations of 29 metabolites [-lg(P) > 3]. Allelic effects were investigated in detail in two markers (RM315 and RM541) as examples.