Spatial metabolomics can reveal intercellular heterogeneity and tissue organization. To achieve highest spatial resolution, we reported a novel Spatial single nuclEar metAboloMics (SEAM) method, a scalable platform combining high resolution imaging mass spectrometry (IMS) and a series of computational algorithms, that can display multiscale/multicolor tissue tomography together with identification and clustering of single nuclei by their in situ metabolic fingerprints. We firstly applied SEAM to a range of wild type mouse tissues, then delineate a consistent pattern of metabolic zonation in mouse liver. We further studied spatial metabolome in human fibrotic liver. Intriguingly, we discovered novel subpopulations of hepatocytes with special metabolic features associated with their proximity to fibrotic niche, which was further validated by spatial transcriptomics with Geo-seq. These demonstrations highlight how SEAM may be used to explore the spatial metabolome and tissue anatomy at single cell level, hence leading to a deeper understanding of the tissue metabolic organization.