We introduce a novel Quantum Walk-based Adaptive Distribution Generator (QW-based ADG) that generates target probability distributions with high precision and efficiency. Our hybrid quantum-classical method integrates variational quantum circuits with discrete-time quantum walks (DTQWs)—specifically, split-step quantum walks (SSQWs) and their entangled extensions—to dynamically tune coin parameters and steer the quantum state evolution. This enables accurate one-dimensional probability modeling for applications such as financial simulation and structured two-dimensional pattern generation, exemplified by MNIST digits. Implemented on the CUDA-Q platform for GPU acceleration, our approach demonstrates significant performance advantages. Benchmarks show the QW-based ADG is not only capable of high-fidelity generation but is also over 10 times faster than a standard Quantum Generative Adversarial Network (QGAN) in 2D tasks while achieving comparable or superior accuracy. These results establish our framework as a powerful and efficient tool that bridges the gap between theoretical quantum algorithms and practical high-performance computing.