Faba bean (Vicia faba L., 2n = 12) is an important food legume crop and widely grown for multiple uses in the world. However, no reference genome is currently available due to its giant genome (approximately 13 Gb) and limited single nucleotide polymorphism (SNP) markers as well as high efficient genotyping tools have been reported for faba bean. In this study, 16.7 billion clean reads were obtained from transcriptome libraries of flowers and leaves of 102 global faba bean accessions. 243,120 unigenes were de novo assembled and functionally annotated. Moreover, a total of 1,579,411 SNPs were identified and were further filtered according to a selection pipeline to develop a high-throughput, flexible and low-cost Faba_bean_130K Targeted Next-Generation Sequencing (TNGS) genotyping platform. A set of 69 Chinese faba bean accessions was genotyped with the TNGS genotyping platform, and the average mapped rate of captured reads to reference transcripts was 93.14%, of which 53.23% were located in the targeted regions. The TNGS genotyping results were validated by Sanger sequencing and the average consistency rate was up to 93.6%. Comprehensive population genetic analyze was performed to the 69 Chinese faba bean germplasm and identified four genetic subgroups correlated with geographic distribution. This study provides valuable genomic resources and a reliable genotyping tool, which could be implemented in genetic and molecular breeding studies to accelerate the faba bean new cultivar development and improvement.