 Neural Radiance Field, NERF, has been successfully used to generate high-quality 3D reconstructions of cultural heritage sites from a single image or video. The paper reviews several NERF methods and compares them against each other. It was found that instant NGP and NERFACTO methods produced the most accurate results, outperforming other methods by a significant margin. Qualitative and quantitative analysis of various datasets showed that NERF methods were able to produce high-quality reconstructions even when dealing with complex textures and shiny surfaces. Furthermore, NERF methods were also able to accurately recreate smaller datasets of lost artifacts. These findings suggest that NERF methods could potentially be used to create detailed 3D models of cultural heritage sites, which would allow for more accurate visualization and communication of these sites. This article was authored by Jimazaka, Jimazaka, Akarami, and others.