CharacterGen is an efficient 3D character generation framework that can generate high-quality and consistent-looking 3D character meshes from a single input image. Using a streamlined generation pipeline and an image-conditioned multi-view diffusion model, it effectively aligns the input pose to a standard form while preserving key attributes from the input image, addressing challenges posed by diverse poses. It also employs a transformer-based universal sparse view reconstruction model and a texture back-projection strategy to generate high-quality texture maps.