How to Use Sora by OpenAI to Generate Videos and Images

How to Use OpenAI’s Sora AI Video Generator

At its core, Sora is OpenAI’s newest tool that turns text into video—think of it as DALL·E meets a movie director. You give it a prompt, and it generates moving images that look like they came straight from a high-end film studio. No need to animate or edit—Sora handles it all behind the scenes.

So how do you actually use it?

Here’s a step-by-step walkthrough, from prompt to playback:

Step 1: Visit Sora and Sign In with Your OpenAI Account

First things first — head over to Sora’s website. Once there, just sign in using your OpenAI account. If you already have a ChatGPT Plus, Team, or Pro subscription, you’re good to go. No extra sign-up needed.

If you don’t have an OpenAI account yet, now’s the time to create one—it’s the key that unlocks all of Sora’s AI magic.

Step 2: Write Your Prompt

To start creating, click on the “Videos” tab or “My media” on the left sidebar. At the bottom of the page, you’ll see a prompt bar that says something like “Describe your video...”—this is where the magic begins.

Tell Sora what kind of video you want by typing a short or detailed description. It could be something simple like:

“a misty mountain at sunrise”

Or something fun and specific like:

“a robot dog skateboarding through Times Square in the rain”

Don’t worry about getting it perfect on the first try—Sora is surprisingly good at interpreting creativity.

Step 3: Upload an Image or Video (Optional)

If you want to guide Sora with a visual reference, you can upload an image or short video clip alongside your text prompt. This helps the AI better understand your vision and generate something more aligned with your idea.

This step is completely optional, but especially useful if you're going for a consistent visual theme, want to match a specific style, or want to base your video on a real-world photo or motion.

Step 4: Customize Your Video Settings

After entering your prompt, you’ll see a variety of settings to shape how your video looks and feels. Here's what you can adjust:

Aspect Ratio: Choose from 16:9 (standard), 1:1 (square), 9:16 (vertical), 3:2, or 2:3—depending on where you plan to share the video.

Resolution:

• 480p – Fastest render time
• 720p – Around 4× slower
• 1080p – Up to 8× slower, available to ChatGPT Pro users only

Duration: Select from 5, 10, 15, or 20 seconds. The higher your plan, the longer your video can be. For example:

• Plus & Team users: up to 10 seconds at 720p
• Pro users: up to 20 seconds at 1080p

Number of Variants: Generate 1, 2, or 4 versions at once so you can pick the best one.

Style Presets: Choose from various artistic styles like:

• film noir v0
• cardboard & papercraft v0
• cartoonify by sora v0
• pixel art by sora v0

...and many more, including some experimental looks like balloon world or archival v0. These can drastically change the mood of your video.

What’s “Storyboard”?

You might notice an option called Storyboard. This lets you create a multi-shot video with several scenes that transition over time. Think of it as piecing together a short narrative or montage. For now, it's still early-stage, but worth exploring if you're aiming to tell a small story in one go.

If you’re just starting out or want quick results, you can skip it for now and focus on single-scene videos.

Step 5: Generate and Review Your Video

Once everything’s set, click the generate button (a small upward arrow icon) and let Sora do its thing. The render time depends on your settings—higher resolution and longer videos take more time.

You’ll get multiple versions (if you chose more than one variant), and they’ll show up in your media panel. You can:

• Click each version to preview
• Compare how different settings affect the outcome
• Select and favorite the best one

Didn’t nail it the first time? Just tweak your prompt, adjust the settings, or try a different preset and generate again. There’s no limit to how many times you can experiment—especially helpful for Plus, Team, and Pro users, who get unlimited generations.

Pro Tip: ChatGPT Pro users also get up to 5 concurrent generations, meaning you can generate multiple videos at the same time—great for speeding up creative workflows.

Step 6: Download Your Video

Happy with the result? You can download it directly from your media panel. Here’s what to expect depending on your plan:

• ChatGPT Plus & Team: Download videos up to 720p. These downloads include a small watermark.
• ChatGPT Pro: Enjoy downloads up to 1080p, without a watermark, and faster generation times overall.

Once downloaded, your AI-generated video is ready to be shared, edited, or archived—whatever you need it for.

Step 7: Upscale or Enhance Your Sora Video (Optional)

While Sora allows you to export up to 1080p resolution (for Pro users), you might want to upscale your video to 2K or 4K, or enhance it with more details—especially if you plan to use it in a professional project or upload it to a platform that favors higher resolution.

That’s where VideoProc Converter AI comes in.

With its AI-powered video enhancer, you can:

• Upscale 720p or 1080p footage to 4K without quality loss.
• Clean up blur, noise, or pixelation from lower-quality generations.
• Recover lost details and sharpen key visuals like faces, textures, or landscapes.

All you need to do is load your Sora video into VideoProc Converter AI, go to the Super Resolution module, choose an AI model and a target resolution (like 4K), and let the AI do the rest.

How to Use Sora to Generate Images

Sora isn’t just about video—it also lets you create still images with the same powerful AI, using simple prompts or reference photos. Here's how to do it:

Step 1: Switch to the Image Generator

Click on the Images tab in the left menu. Alternatively, go to My Media and switch the content type from “Video” to “Image” in the input box at the bottom.

This changes the interface slightly, but the basic workflow is the same.

Step 2: Enter Your Prompt

Just like with video generation, start by typing what you want to see. Your prompt can be anything from “a cyberpunk city at night” to “a watercolor-style painting of a fox in a snowy forest.”

The clearer or more creative your input, the better the results.

Step 3: (Optional) Upload an Image

You can also guide the style or structure by uploading a photo. This is especially helpful if you want Sora to transform or reference a specific subject or composition.

Step 4: Adjust Your Settings

Before generating your image, tweak a few options to suit your needs:

• Aspect Ratio: Choose from 2:3, 3:2, 1:1
• Number of Variants: Generate 1, 2, or 4 versions at once
• Presets: Add stylistic flair with options like Film Noir, Cartoonify, Pixel Art, or Balloon World

When you're ready, hit the Create Image button (look for the upward arrow).

Step 5: Review and Remix

Once your image is generated (this may take up to 2 minutes depending on prompt complexity), you can click to view it in full.

If it’s not quite what you imagined—or if you want to explore a variation—select Remix or press R on your keyboard to describe changes. Sora will generate a new image based on the original.

Step 6: Download Your Image

Click the download icon to save your favorite result to your device.

Step 7: Upscale or Enhance Your Image (Optional)

If you want to polish the final image—for example, for printing, commercial design, or high-res display—you might want to take it a step further. In this case, you can try Aiarty Image Enhancer.

With Aiarty, you can upscale low-res AI art to 4K/8K and even 32K, remove unwanted blur or jagged edges, refine textures, sharpen faces, and restore clarity.

Just upload your Sora-generated image into Aiarty, pick the AI model that best fits your image type, and let the AI handle the details.