SatSynthBurst is a synthetic benchmark dataset for multi-image super-resolution of satellite imagery, introduced alongside the SuperF: Neural Implicit Fields for Multi-Image Super-Resolution paper. It is designed to support the evaluation of methods that reconstruct a high-resolution image from multiple low-resolution observations with sub-pixel shifts. The dataset is derived from the WorldStrat dataset and consists of simulated satellite image bursts that reflect the characteristics of real remote sensing acquisitions, including repeated low-resolution observations of the same scene. The applied sub-pixel shifts are known and controlled, enabling precise evaluation of alignment performance in addition to reconstruction quality. The dataset is intended for benchmarking reconstruction quality, alignment robustness, and generalization in burst super-resolution settings.