Sora, the upcoming Text to-video model from OpenAI. Sora can generate videos up to a minute long while maintaining visual quality and adherence to the user’s prompt.
OpenAI Sora possesses the ability to generate complete videos in one go or extend existing videos to increase their length. This is achieved by providing the model with foresight over multiple frames simultaneously, overcoming the challenge of maintaining continuity when a subject briefly exits the frame. Utilizing a transformer architecture similar to GPT models, Sora demonstrates exceptional scalability performance. In Sora's framework, videos and images are represented as collections of smaller units known as patches, which are analogous to tokens in GPT. This unified approach to data representation allows diffusion transformers to be trained on a broader spectrum of visual data than previously feasible, encompassing varying durations, resolutions, and aspect ratios.
Who can get access to Sora AI:
Sora is now being made accessible to red teamers for evaluating crucial areas for potential harms or risks. Additionally, access is being extended to a select group of visual artists, designers, and filmmakers, enabling them to provide insights on how to enhance the model to best serve creative professionals.
Sora is now being made accessible to red teamers for evaluating crucial areas for potential harm or risks. Additionally, access is being extended to a select group of visual artists, designers, and filmmakers, enabling them to provide insights on how to enhance the model to best serve creative professionals.
The Future is Intriguing and Dynamic:
Sora is now being made accessible to red teamers for evaluating crucial areas for potential harm or risks. Additionally, access is being extended to a select group of visual artists, designers, and filmmakers, enabling them to provide insights on how to enhance the model to best serve creative professionals with caution, ensuring that they serve the collective good and contribute positively to shaping a better future for all.
*Share your Opinion Guys!!
*
#openai #sora #texttovideo #ai #gpt #dalle
Source: https://openai.com/sora
Top comments (0)