PhotoVerse is a no-adjustment image customization method that utilizes a text-to-image diffusion model, empowering users to create customized images based on specific concepts and prompts. Compared to existing methods, PhotoVerse boasts advantages such as fast generation speed, high-quality images, and identity preservation. It employs a dual-branch conditional mechanism, achieving effective control over the image generation process in both text and image domains. Furthermore, it introduces facial identity loss as a new component during training to enhance identity retention. PhotoVerse relies solely on a single facial photo of the target identity, eliminating the need for time-consuming adjustments, significantly reducing the resource cost of image generation. After a single training, our method can generate high-quality images within seconds. Additionally, our method can generate diverse images encompassing various scenes and styles.