Accurate electrochemical-thermal simulation of lithium-ion batteries (LIBs) is critical for improving performance, ensuring safety, and extending service life. However, conventional pseudo-two-dimensional (P2D) models often fail to capture in-plane inhomogeneities, particularly in large-format pouch cells, due to geometric simplifications. While pseudo-four-dimensional (P4D) models can resolve such effects, their practical use is limited by the high computational cost associated with nonlinear iterative solution procedures. This study introduces a non-iterative electrochemical–thermal solution framework based on a pseudo-four-dimensional (P4D) modeling approach, which incorporates full three-dimensional (3D) spatial resolution alongside a pseudo-dimension representing intra-particle lithium diffusion, enabling high-fidelity analysis of internal gradients in large-format LIBs. To mitigate the computational burden of P4D simulations, an efficient non-iterative numerical scheme is developed by combining staggered time integration, Taylor-series-based linearization, and an adaptive time-stepping strategy guided by output gradients. The governing equations are reformulated into a sequence of linear subsystems, allowing the coupled system to be solved without nonlinear iterations. Validation against experimental discharge data and commercial finite element simulations demonstrates that the proposed method achieves comparable accuracy while reducing computational time by up to 5-fold. The proposed method is further applied to analyze spatial inhomogeneities in large-format pouch cells under varying aspect ratios and charging conditions. Results show that spatial gradients in electrolyte concentration, electrode overpotential, and interfacial reaction rates become increasingly pronounced with larger cell geometries and higher C-rates. These in-plane non-uniformities significantly influence voltage dynamics and remain unresolved in lower-dimensional models such as the P2D model. In particular, the P4D framework enables precise identification of localized regions susceptible to lithium plating–insights inaccessible in conventional modeling approaches. Overall, the results demonstrate that the proposed non-iterative framework enables efficient and scalable high-fidelity P4D simulations, providing a practical pathway for the design and analysis of large-format lithium-ion batteries.