OpenAI Sora 2 Makes a Splashing Debut with Its First - Ever App Launch
5 day ago / Read about 0 minute
Author:小编   

On September 30, 2025, during a live - streamed event, OpenAI made an official announcement, introducing its latest - generation video generation model, Sora 2. This move places it in direct competition with Google's Veo 3.

Sora 2 stands out by incorporating audio - video synchronous generation technology. For the very first time, this model enables precise alignment between visuals and various audio elements, including sound effects, object interaction noises, and dialogue. In terms of physical simulation, it has achieved a remarkable leap in accuracy. It can now accurately model complex dynamic scenarios, such as the intricate movements in gymnastics and the mechanics involved in diving.

Moreover, Sora 2 brings about significant advancements in multiple aspects. It supports breakthroughs in image resolution, allowing for the creation of videos with higher - definition visuals. It also offers enhanced style manipulation capabilities, enabling users to easily adjust the artistic style of the generated videos. Additionally, it adheres more closely to user instructions.

Alongside the model, OpenAI has launched the iOS app Sora. One of its standout features is 'Join the Scene,' which empowers users to insert themselves or their friends into the generated videos, adding a personalized touch.

OpenAI has also revealed that the Sora app API will be made accessible to third - party developers within a few weeks. Initially, ChatGPT users in the United States and Canada will have the privilege of being the first to experience this new offering. Subsequently, it will be gradually rolled out to other regions.

Sora 2 is available to all users free of charge, but there is a cap on the number of uses. For those seeking higher - quality results, ChatGPT Pro users can gain access to the 'Sora 2 Pro' model.