Hemophilia A is caused by a deficiency in the factor VIII (FVIII) gene. Constrained by limited packaging capacity, even the 4.3-kb B domain-deleted FVIII remained a challenge for delivery by a single adeno-associated viral (AAV) vector. Studies have shown that up to a 6.6-kb vector sequence may be packaged into AAV virions, which suggested an alternative strategy for hemophilia A gene therapy. To explore the usefulness of AAV vectors carrying an oversized FVIII gene, we constructed the AAV-FVIII vector under the control of a beta-actin promoter with a cytomegalovirus enhancer (CB) and a bovine growth hormone (bGH) poly(A) sequence. The CB promoter plus bGH signal was shown to be 3- to 5-fold more potent than the mini-transthyretin (TTR) promoter with a synthetic poly(A) sequence for directing FVIII expression in the liver. Despite the 5.75-kb genome size of pAAV-CB-FVIII, sufficient AAV vectors were produced for in vivo testing. Approximately 3- to 5-fold more FVIII secretion was observed in animals receiving AAV-CB-FVIII vectors than in those receiving standard-sized AAV-TTR-FVIII vectors. Both the activated partial thromboplastin time assay and the whole blood thromboelastographic analysis confirmed that AAV-FVIII vectors fully corrected the bleeding phenotype of hemophilia mice. These results suggest that AAV vectors with an oversized genome should be useful for not only hemophilia A gene therapy but also other diseases with large cDNA such as muscular dystrophy and cystic fibrosis.