Hailuo 2.3 Fast

The Hailuo 2.3 Fast is an accelerated video generation model developed by MiniMax, designed to provide high-speed, cost-effective output while maintaining cinematic visual quality. Released in October 2025 as part of the Hailuo 2.3 family, it specifically targets the image-to-video (I2V) workflow, offering generation speeds up to 50% faster than the standard version. It is built to serve creators and developers who require rapid iteration and high-volume production for social media, advertising, and concept previews.

Technically, the model builds upon the Noise-aware Compute Redistribution (NCR) architecture introduced in previous iterations, which optimizes computational resource allocation during the diffusion process. This foundation allows Hailuo 2.3 Fast to deliver improved physical realism and character consistency. It excels at rendering complex human motions—such as dancing, gymnastics, and intricate choreography—with higher fluidity and fewer limb distortions than earlier models.

The model supports video generation at 768p and 1080p resolutions, with clip durations typically ranging from 6 to 10 seconds. It is particularly noted for its ability to capture nuanced character performances, including micro-expressions and subtle facial gestures. In addition to photorealistic styles, Hailuo 2.3 Fast maintains stability across various artistic aesthetics, including anime, game CG, and traditional ink-wash painting styles.