Deepseek V4.1 Flash Release Video [Made with Deepseek V4.1 Flash]

AI 摘要

A Reddit user benchmarked the new Deepseek V4.1 Flash model using motion videos. The user noted that the Deepseek V4.1 Flash model has significantly improved, graduating from a "Luna class model to nearly an Opus class model" based on its performance with motion videos. This assessment highlights the model's enhanced capabilities in processing and generating motion-related content.