Wheat is a staple crop of global relevance that requires further improvement in grain yield to meet the increasing demands of the estimated population growth by 2050. Grain yield is mainly determined by two usually negatively correlated components, grain number and grain weight. Thus, the trade-off between the two yield components is a bottleneck in an attempt to increase grain yield. In this study, a QTL mapping was conducted for several grain yield-related traits using a F2 population derived from a cross between a pyramidale wheat accession TN28 and a durum wheat accession Langdon. In genotype investigation, genome-wide genetic polymorphisms of 144 F2 individuals were obtained by an effective high-throughput sequencing method MIG-seq. 11 grain yield-related traits were obtained and among these the grain number per spikelet was investigated in a detailed way where the spike was divided into three parts (i.e., grain number per spikelet at apical, central and basal parts of the spike). QTL analysis of the 11 traits were performed by the method of Interval Mapping. Two QTLs contributing to grain yield were detected on chromosome 7A and 7B. The QTL on chromosome 7B was identified for grain number per spikelet at the central part of the spike and in a whole position of spike. Remarkably, the QTL on chromosome 7A was detected for increasing both grain number and grain weight. It was identified for grain number per spikelet at the apical and central part of the spike, and in a whole position of spike, which was also identified for thousand grain weight. This QTL for grain number per spikelet at the apical part of the spike showed a significant high LOD score of 9.09. Its additive effect was 0.26 and it can explain up to 28.5% phenotypic variance. On the other hand, this QTL for thousand grain weight was detected with a LOD score of 4.00. Its additive effect was 3.88 and it can explain up to 14.3% phenotypic variance. The associations between the two traits and the QTL-linked marker showed that TN28 alleles cause not only higher grain number but also higher thousand grain weight. A further targeted exploitation of the available variation, coupled with advanced genomic approaches, may help better understand the genetic basis of yield-related traits and provide valuable resources for breeding high grain yield in wheat.