OpenAI has officially launched its generative video model Sora. Initially introduced in February 2024 as a research preview, Sora is now available to the general public through ChatGPT Plus and Pro subscriptions. This release coincides with OpenAI’s “12 Days of OpenAI” campaign, which features daily product launches and demonstrations.
Sora operates similarly to the DALL-E image generation model, allowing users to create video clips by providing text prompts. The model supports advanced features such as generating videos from still images, extending existing videos, and filling in missing frames.
Users can create videos with resolutions up to 1080p and durations of up to twenty seconds. Supported aspect ratios include widescreen, vertical, and square formats. Sora also features a storyboard tool, enabling users to customize input frame by frame and integrate their own assets for remixing or blending content.
The platform offers “Featured” and “Recent” feeds showcasing user-generated content, fostering a creative community. ChatGPT Plus subscribers can create up to fifty videos per month at 480p resolution or less at 720p. Pro subscribers gain access to higher resolutions, longer durations, and significantly increased usage limits.
Sora is currently available at sora.com for ChatGPT Plus and Pro users in 137 countries, excluding the EU and the UK, with plans for broader access and tailored pricing models in 2025. The primary audience includes content creators, marketers, educators, and anyone exploring innovative storytelling methods.
