Chemometric validation of the reported experimental activities of nicotinamide and carboxamide analogues against diabetes were determined via quantitative structure-activity relationship (QSAR) model using ATS1p, AATSC1m, nHother and RDF80p descriptors. Meanwhile, GPBAR_HUMAN with a UniProtKB code (Q8TDU6) for G protein-coupled bile acid was known as a potential drug target in human TGR5 of Type 2 diabetes mellitus (T2DM) whose crystal structure wasn’t found in the Protein Data Bank. With the aid of in-Silico approach via homology modeling helps to build a 3D protein (target) model with the availability of GPBAR_HUMAN sequence data. The swiss-Model online workspace was used to model the GPBAR_HUMAN receptor using the Cryo-EM structure of the INT-777-bound GPBAR-Gs complex with PDB code (7CFN) as a potential template. The built receptor was evaluated, validated, and used as a target protein for molecular docking simulation of some potent nicotinamide and carboxamide series. The virtual screening studies via molecular docking revealed the binding pockets, binding modes and poses of the GPBAR_HUMAN receptor with some prominent anti-diabetic agents. Analysis of these interactions led to computational design via a structure-based approach of new potent anti-diabetic compounds with significant binding scores and better interactions with the GPBAR_HUMAN receptor. Validation of the designed compounds via Drug-likeness and ADME prediction confirmed that these compounds are orally bioavailable with a good lipophilicity index for lipid environments and show zero violation to the drug assessment rules in order to be considered as drug candidates. Therefore, in-vitro and an in-vivo test to transform these findings into potent therapeutics are strongly recommended.