Recraft V3 is the most recent model and is used by default. Released in October 2024, it ranked first for five consecutive months on the Artificial Analysis public benchmark on Hugging Face, maintaining a clear lead over models from Midjourney, OpenAI, and others. V3 introduced major advances in photorealism and text rendering. It was the first Recraft model to generate mid-size text accurately and, as of 2025, is the only model capable of placing text at specific positions in an image. It supports a rich library of curated styles, an infinite library of AI-generated styles, and—like V2—can maintain style consistency from a small set of example images without retraining. V3 also includes the Artistic level parameter for adjusting the artistic fidelity of an image. Lower values produce more standard images, such as direct portraits, while higher values create less conventional compositions with more peculiar angles or perspectives. Because of its flexibility and quality, V3 is recommended for most use cases.