MiniMax Image-01 is a high-performance text-to-image generation model developed by the Chinese AI laboratory MiniMax. Officially launched on February 28, 2025, the model is engineered to provide cinematic-quality visuals with high prompt fidelity and realistic rendering. It serves as a core component of MiniMax’s multimodal ecosystem, complementing their Hailuo AI video models and the MiniMax-01 large language model series.
The model is optimized for precise prompt adherence, enabling it to interpret complex natural language descriptions with high accuracy. It demonstrates advanced proficiency in visual composition, particularly in the interplay of light and shadow, which contributes to the depth and atmosphere of generated scenes. Image-01 is specifically noted for its ability to render lifelike human subjects—preserving natural skin textures and expressions—and authentic material details in inanimate objects.
Image-01 supports a broad range of creative formats, offering seven standard aspect ratios including 1:1, 16:9, and 21:9. For professional workflows, it provides high-batch efficiency, allowing users to generate up to nine variations in a single request. The system includes a built-in prompt optimizer to automatically enhance user input and supports high-resolution outputs up to 2048x2048 pixels. Beyond text-to-image, the model includes image-to-image capabilities with support for character reference, ensuring consistent appearance across different scenes.