Abstract The lateral broadband fracturing (LBF) technique for vertical well in tight oil reservoir has been proven to provide much larger stimulated reservoir volume (SRV) by widening its width. Although this new fracturing technique has been successfully applied to the development of tight oil reservoirs in Ordos Basin, China, there is still a lack of models and methods to characterize the imbibition of matrix-fracture system, which is heterogeneous permeability distribution. In this paper, a multilinear fractal model considering Nuclear Magnetic Resonance-based research on imbibition (MFMI) is established to characterize the flow characteristics of lateral broadband fracturing vertical wells (LVWs) in tight oil reservoirs by combining the dual-porosity fractal model considering imbibition and the quad-linear flow model. Due to the application of LBF, the nonuniform distribution of fracture system is characterized by fractal porosity and fractal permeability. In addition, the imbibition in SRV of tight oil reservoir is quantitatively characterized by Nuclear Magnetic Resonance (NMR) data. And the production performance of LVWs is quantified by the MFMI. By using the Laplace transformation, Bessel function, iteration, and Stehfest numerical inversion algorithms, the approximate analytic solutions of our established model, including primary hydraulic fractures, SRV, and unstimulated reservoir volume (USRV), are derived. The solutions of pressure and production are used to compare and analyze in order to discuss the influence of parameters related to lateral broadband fracturing (such as fractal parameters, reservoir parameters, and width of SRV) on flow behavior of LVWs in a tight oil reservoir. The modeling results show that the fractal parameters of fracture system have great effect on the fluid flow in LVWs, and LBF contributes to imbibition production.