Kling AI is a diffusion transformer text-to-video model that was created by the Chinese company Kuaishou Technology and later announced on June 6, 2024. Kuaishou claimed that the model can generate up to two minutes of video at 30 frames per second and in 1080p resolution. The large language model (LLM) uses three-dimensional face and body reconstruction using the company's proprietary 3D VAE (so-called "3D spatiotemporal joint attention mechanism"[1]) and the user can create videos in various aspect ratios.[2]

Following KwaiYii and Kolors,[3] the model's supposed video output was showcased on their website, and as of June 2024, the model is in beta access.[4] The website also showcases a supposed demo of the "3D face and body reconstruction" technology's manipulation of a whole-body photo.[5]

The model has been compared to that of OpenAI's Sora text-to-video model; Kling AI has been deemed to be the "rival to Sora".[6][7][8]

References

edit
  1. ^ "可灵大模型" [Kling Large Model]. kling.kuaishou.com (in Chinese (China)). Archived from the original on June 8, 2024. Retrieved June 9, 2024.
  2. ^ Dwivedi, Aryan (June 8, 2024). "China's New Text-to-Video Model 'Kling' offers Tough Competition To Open AI's Sora". India.com. Archived from the original on June 9, 2024. Retrieved June 9, 2024.
  3. ^ Chen, Wency (June 7, 2024). "Chinese short video app Kuaishou unveils Sora-style product amid AI race". South China Morning Post. Archived from the original on June 9, 2024. Retrieved June 9, 2024.
  4. ^ Leung, Tony (June 8, 2024). "Kling AI | Sora like Text-To-Video Model". Medium. Retrieved June 9, 2024.
  5. ^ Eliaçık, Eray (June 7, 2024). "What Is Kling AI Video Tool? Explained With Examples". Dataconomy. Archived from the original on June 7, 2024. Retrieved June 9, 2024.
  6. ^ Pandey, Nikhil (June 7, 2024). "China's Text-To-Video AI Tool, Dubbed Sora Killer, Blows Everyone's Mind". New Delhi Television. Archived from the original on June 9, 2024. Retrieved June 9, 2024.
  7. ^ "OpenAI Sora's biggest rival is here: Chinese AI model Kling creates videos with more accuracy". The Indian Express. June 7, 2024. Archived from the original on June 9, 2024. Retrieved June 9, 2024.{{cite web}}: CS1 maint: bot: original URL status unknown (link)
  8. ^ Cong, Guan; Wei, Han (June 7, 2024). "Kuaishou Unveils a Rival to Sora That Produces Video From Text - Caixin Global". Caixin Global. Archived from the original on June 7, 2024. Retrieved June 9, 2024.