How to Use Sora for AI Video Generation in 2025

The AI video generation world is moving insanely fast, and it can be hard to keep up. Every other week, there’s a new tool, a new model, or a “game-changing” feature update, and most of them require a paid subscription just to try. That makes it unrealistic to test everything out yourself!

But you’ve definitely heard of, or even tried ChatGPT. The same company behind it, OpenAI, also has an AI video model— Sora.

In this article, we’ll show you how to use Sora, from the basics like text-to-video, image-to-video, and video-to-video generations, to more advanced tools like Storyboard, Remix, and Loop. You’ll get a clear picture of what it can do, so you can decide if it’s worth exploring or just another flashy AI distraction.

What is Sora

Sora is OpenAI’s video generation model that can turn text, images, or videos into new video content. Unlike its image generation model, DALL·E, which can be accessed directly through the ChatGPT interface, Sora is cloud-based and is only available through its own webpage (as of August 2025).

You can use it in three main ways:

• Text-to-video : Describe what you want in your video using words.
• Image-to-video : Upload a photo or use an AI-generated image from Sora and turn it into a short video (either automatically or by adding a text prompt to guide the results more precisely).
• Video-to-video: Combine two videos into one seamless clip (similar to morphing).

Whichever input you use, you can style your videos using available presets like noir, cartoon, pixelated, etc., or even create your own.

How to Use Sora Ai Video Generation

Text-to-Video Generation

Text-to-video is the easiest way to start creating videos with Sora. You don’t need to prepare any assets. Just type what you want in your video, and Sora will generate it. Here’s how to do it:

Go to Sora’s homepage and describe the scene you want in the input box:

🐱 Meow Memo: Be as specific as possible with your prompt. Mention details like camera angle, lighting, colors, movement, and mood to help Sora better understand the kind of result you’re aiming for.

Different settings can be found at the bottom of the input box, including:

• Aspect Ratio : 16:9, 3:2, 1:1, 2:3, 9:16
• Duration : 5s, 10s, 15s, 20s
• Resolution : 480p, 720p, 1080p
• Variants : Choose up to 4 versions to generate
• Preset : Choose from Archival, Film Noir, Cartoonify, etc.

Before you submit the prompt, make sure “Video” is selected:

Once you’ve submitted your prompt, your video will be queued for processing. You can click the bell icon to track its progress and access the finished video there:

…or go to “My media” to view your result:

Clicking on the video opens up more options. You can add it to a storyboard (View Story), or use tools like Recut, Remix, Blend, and Loop to further modify the video:

Image-to-Video Generation

With Sora, you can bring any image to life. It can be one you generated in Sora, uploaded yourself, or even one you found on the Explore page.

Animating an Image from Your Library or Device

To start, click the add icon in the input bar to upload an image or choose one from your library:

Then, enter a text prompt to describe what you want in the video and adjust other settings as needed:

🐱 Meow Memo: You can let Sora take full creative control by animating an image without adding any text prompt.

Animating an Image from The Explore Page

If you want to animate an image from the “Explore” page, just click on the one you like:

Then click “Create video” :

Just like before, you can choose to add a text prompt for more control, or skip it and let Sora animate the image on its own:

Video-to-Video Generation

Besides generating videos from scratch, Sora can also blend two clips into a new one. The Blend feature lets you merge the content and style of one video with another. You can control how much influence each video has using one of the blend curves available:

• Transition : Seamlessly flow between clips
• Sample : Apply the style of one clip onto another
• Mix : Merge both clips together
• Custom : create your own blend flow manually

To create a blend video, just click on one of your existing videos, then select “Blend” to add another video:

Once you’ve added the two videos you want to blend, you can trim them by dragging the edges of the box below:

You can also adjust the blend curve by picking one of the preset options:

…or by customizing it yourself by dragging the curve:

Depending on the blend curve you choose, this feature can be great for creative scene transitions, character morphs, storytelling effects, or just experimenting with different visuals.

Here’s the result for the screenshot above using the “Mix” blend:

More Than Basic AI Video Generation: Advanced Sora Video Editing

Storyboard: Extend and Arrange Your Videos

The Storyboard feature lets you arrange and edit your video frames in a custom sequence on a timeline, like using a typical video editor. You can drag, rearrange, or delete clips to shape your narrative however you want.

For each card (or frame), you can upload a video, add an image, or write a text prompt to describe what should happen at that moment:

You can access it by clicking the “Storyboard” option below the input section, or by clicking “Edit story” at the bottom when viewing your video.

🐱 Meow Memo: Leaving enough space between cards makes it easier for Sora to connect them smoothly.

Remix: Replace, Remove, or Edit Elements

The Remix tool lets you edit your video using text prompts. You can add, remove, or modify objects in your video, and choose how significant you want the changes (strong, mild, subtle, or custom).

Loop: Trim and Create Seamless Video Loops

Loop lets you turn any part of your video into a smooth, seamless loop. Just open a video from your library, click “Loop” , and use the slider to select the section you want to repeat. Sora will automatically blend the start and end frames to create a looping effect.

Is Sora Free

As of August 2025, Sora is only available for ChatGPT Plus and Pro users. Pricing starts from $20/month:

Final Thoughts

Sora’s AI video generation produces impressive results, and the Storyboard feature is one of its biggest selling points. It makes it easy to create individual scenes and piece them together into a full story that can be up to 20 seconds long. On top of that, the Remix tool allows you to make changes and refine the video using text prompts until you’re satisfied.

However, the movements can be glitchy sometimes, and it might misinterpret parts of your prompt. You might need to refine it a few times to get the result you want, and that can burn through your credits fast.

Even if it’s not perfect, it’s easily one of the best AI video tools out there right now. If you’re already on ChatGPT Plus or Pro, there’s no reason not to test it out. And if you’re not, it’s worth testing to see what it can create if you have the budget.