• Source:JND
Add as a preferred

OpenAI Sora: ChatGPT maker OpenAI announced its text-to-video generation model Sora, which can generate videos that can go up to one minute in length while respecting a user's prompt. The company recently opened access to the model to red teamers and select users to gain feedback and gauge the inherent risks. As the model continues to gain buzz among the tech fraternity, OpenAI CEO Sam Altman shared some results generated using prompts on X (formerly Twitter).

These results were based on the prompts received by users on the microblogging platform. One of the results was based on the prompt received by CRED CEO Kunal Shah. "A bicycle race on ocean with different animals as athletes riding the bicycles with drone camera view," @kunalb11 posted. A few hours after the comment, Altman shared a video generated using Sora.

Also Read: OpenAI's New AI Model Sora Can Create One Minute Videos Based On Your Text Prompts; Details

Here is the output for your reference:

Soon after Altman (@sama) posted: "We'd like to show you what Sora can do, Please reply with captions for videos you'd like to see and we'll start making some [edited capitalisation]," several users shared their request in the comments box on X. These responses ranged from minting a wizard, half duck half dragon creature, golden retrievers podcasting, futuristic city, drone race on Mars and more. Here are a few responses.

1.

2.

3.

4.

5.

6.

Through Sora's testing, the company is currently working to gather feedback on the output. "Sora is able to generate complex scenes with multiple characters, specific types of motion, and accurate details of the subject and background," OpenAI added. While it can generate compelling output, it may struggle to cope with complex scenes, the company noted.

Also Read: ChatGPT Gets Human-Like Traits: Can Now Hear, See And Speak In Humanly Language; How To Use

Apart from generating videos from textual information, Sora can also animate existing images to process a video out of them. Plus, it can also extend or fill in frames if they are missing in a video. You can read about the model here.