OpenAI recently launched a 12-day marathon of their latest technologies, capturing the attention of the tech world. Among the most eagerly awaited announcements was SORA, a groundbreaking model designed to transform text into video.

The anticipation for this technology has been building for nearly 10 months since its first introduction. But has SORA managed to meet the high expectations? While it has sparked significant interest, the model has also faced criticism for certain shortcomings. As noted by The Atlantic, SORA has aroused enormous interest among the community, but at the same time has become the subject of numerous discussions.

In this article, we will explore SORA’s key features and assess how well it delivers on its ambitious promises.

What is SORA?

SORA is a new model from OpenAI that allows you to create videos based on text descriptions. It uses artificial intelligence to generate realistic scenes that previously required a team of designers and a lot of time.

The main idea of ​​SORA is to make video creation accessible to everyone. For example, you can describe: "A small dog runs through a field of flowers at sunset," and in a few minutes you will get a video that matches this description.

Here are some analogues that already exist on the market:

• Runway generates short videos, but with a lower level of detail.
• Kling AI is a video creation tool with a focus on realistic animation.
• Hailou AI is a model that specializes in generating videos for business purposes, such as advertising or presentations.

SORA is different in that it offers frame-by-frame control and the ability to integrate your own photos or videos for editing. This makes it unique among similar tools.

Criticism of SORA

The release of SORA brought not only excitement but also a wave of criticism. As noted by The New York Post, some generated videos look too “artificial” and contain visual flaws. For instance, textures can appear blurry, and character movements might seem unnatural. This disappointed users who expected higher quality.

The Verge highlighted another issue — the complexity of using the tool. While SORA offers many features, such as frame-by-frame editing, these require time and patience, especially for beginners. Another drawback is the cost — the SORA subscription is considered expensive by many users.

But is it really that bad? We believe these flaws don’t mean the first release was a failure. SORA already shows great potential. The developers are listening to feedback and actively working on improvements. While the model does have its issues, they will be gradually minimized with future updates. This is just the beginning, and SORA has a bright future ahead.

How to Work with SORA

To use SORA, you’ll need an active ChatGPT Plus ($20/month) or ChatGPT Pro ($200/month) plan. 

At the bottom of the screen, you’ll find a text input field. If you’d like to start with an existing image or video, click the «+» icon in the input field to upload your file. 

 Before generating your video, check the following parameters:

• Aspect ratio (16:9, 1:1, 9:16),
• Resolution,
• Duration,
• Number of variations.

These settings will affect the cost of the generation. Hover over the «?» icon next to the settings to see the required credits.

Click «Submit», and SORA will process your prompt. The process takes up to one minute. You can track the status in the upper-right corner of the screen.

Key Features

One of the standout features of SORA is Storyboard. This tool allows you to plan your video frame by frame, specifying exactly what should appear and when.

It’s simple: you create a script for your video and add precise instructions. For example, in our video, we specified that a seagull should appear at the 3-second mark, flying over the beach. 

The result? The seagulls did appear, but not exactly at the specified moment. However, the fact that they showed up and fit well into the scene’s atmosphere demonstrates that the tool works quite effectively.

Another interesting feature is Blend, which allows you to merge different videos and create seamless transitions between scenes. We tried combining two videos of dogs. 

To be honest, our example might not look perfect, but the feature performed really well. With Blend, you can experiment and create videos that feel cohesive and natural.

They also introduced a variety of exciting features that open up new possibilities for video creation: Remix, Recut, Loop, and Style Presets.

• Remix. This feature allows you to modify scenes by changing backgrounds, replacing objects, or adding and removing elements.
• Recut. With Recut, you can shorten or extend video frames, achieving the perfect edit.
• Loop. Creates seamless animations that repeat smoothly.
• Style Presets. Instantly apply a predefined visual style, such as a papercraft look, without the need for extensive prompt editing (take a look at the image below).

Final Word

SORA is a tool that truly impresses with its capabilities. Yes, there are still some challenges: the servers are often overloaded due to the high number of users eager to test the model. But this is understandable — the interest in the tool is enormous.

Despite these difficulties, the overall impression is largely positive. SORA already offers many exciting features that expand creative possibilities. Over time, as the developers stabilize the servers and address user feedback, this tool has every chance to become a favorite among creative professionals.