Extrachromosomal circular DNAs (ecDNAs) have emerged as important intra-cellular mobile genetic elements that affect gene copy number and exert in trans regulatory roles within the cell nucleus. Here, we describe scCircle-seq, a method for genomically profiling ecDNAs and unraveling their diversity and complexity in single cells. We implemented and validated scCircle-seq in normal and cancer cell lines, demonstrating that most ecDNAs largely vary between cells and are stochastically inherited during cell division, although their genomic landscape is cell type-specific and can be used to accurately cluster cells of the same origin. ecDNAs are preferentially produced from chromatin regions enriched in H3K9me3 histone mark and are induced during replication stress conditions. Concomitant sequencing of ecDNAs and RNA from the same cell uncovered the absence of correlation between ecDNA copy number and gene expression levels, except for few oncogenes, including MYC, contained within a large ecDNA in colorectal cancer cells. Finally, we applied scCircle-seq to one prostate cancer and two breast cancer specimens, revealing different ecDNA genomic landscapes and, in one triple negative breast cancer sample, the presence of two distinct tumor subclones harboring distinct ecDNA signatures. scCircle-seq is a scalable tool that can be used to dissect the complexity of ecDNAs across different cell and tissue types, and further expands the potential of ecDNAs for cancer diagnostics.