Sora 2
OpenAI's flagship video model — up to 25 seconds of continuous video with realistic physics and the Cameo feature for real-world character insertion.
Max Duration
30 seconds
Resolution
720p, 1080p
Aspect Ratios
16:9, 9:16, 1:1
What makes Sora 2 unique
Up to 25 seconds of continuous video on Pro — the longest single-clip duration of any major model
Cameo feature — inject real-world people into generated environments while preserving their appearance and voice
Industry-leading physics simulation: realistic rebounds, gravity, momentum, and object interactions
Strong style control for cinematic, animated, Pixar-style, and artistic looks
Smooth camera transitions, tracking shots, and depth-of-field effects
Diffusion Transformer architecture treats video as sequences of visual patches over time
Best for
How Sora 2 works
Sora 2 is a Diffusion Transformer that treats video data as sequences of visual patches over time — analogous to how LLMs process text tokens. This enables understanding of spatial details within frames and temporal relationships between frames, producing videos with consistent motion, realistic physics, and smooth camera work.
How to use Sora 2 on Flashloop
Select Sora 2
Open Flashloop and choose Sora 2 from the model selector in the video creator.
Enter your prompt
Describe what you want to see in your video — be as detailed as you like.
Generate & download
Hit generate and your video will be ready in seconds. Download or share directly.
Ready to create with Sora 2?
Start generating videos with Sora 2 on Flashloop — available on iOS, Android, and web.
Start Creating with Sora 2