Beijing, China - October 2, 2025 - The latest 2.5 Turbo Video model from Kling AI, the world leading AI-powered creative platform, today tops the video arena in Artificial Analysis, an industry-recognized benchmark that ranks video generators via public voting.
Kling AI’s 2.5 Turbo Video model outperformed Google’s Veo 3 and Luma Labs’ Ray 3 to secure the No.1 spot in both the text-to-video and image-to-video leaderboards, according to the latest results released by Artificial Analysis. The benchmark ranks the video generation models based on user preferences in blind comparisons.
Kling AI’s 2.5 Turbo Video model, released less than a week ago, supports 5-second and 10-second video generation at resolutions up to 1080p. It represents a significant leap forward from its predecessor, Kling 2.1, with major improvements in prompt adherence. The new model also excels at generating highly dynamic motion and complex camera movements that were previously difficult to achieve.
Available at 25 credits per a 5-second video, the model features a a close to 30% price cut in comparison to its predecessor 2.1, signaling Kling AI’s commitment to empowering content creators globally to better stories with the power of AI.
Since its launch in June last year, Kling AI has undergone over 30 iterations and has powered the generation of more than 200 million videos and 400 million images, supporting a community of over 45 million creators.
For more, click here to view the announcement from Artificial Analysis and the model release note from Kling AI.
Media Contact
Company Name: KLING AI
Contact Person: Jack Huang
Email: Send Email
City: Beijing
Country: China
Website: https://klingai.com/