OpenAI's Sora vs Other AI Video Generators
Sora vs Leonardo, Luma, Runway, and Kling
I’ve been excited about the release of Sora for a long time — ever since the first leak was released many months ago.
But since the release of Sora was so delayed — many other amazing AI video generators, such as Leonardo, Luma, Runway, and Kling, have already been launched.
These AI video generators were used to create the infamous Coca-Cola commercial.
If you haven’t watched it — you should watch it now.
For a real comparison of Sora, I wanted to recreate some AI-generated scenes in the commercial.
The first scene I decided to test using Sora — is a simple one of the trucks driving towards the screen.
👉 Sign up for our free 5-day email course to grow 🚀 and earn💲👈
Scene 1 — Coca Cola Trucks driving Towards the Camera
I uploaded this image to ChatGPT o1 and asked it to make me a prompt to recreate this scene.
My prompt
Can you write a prompt for me for an ai video generator to describe what is on this screenshot? I want to use this prompt to generate an AI video using this as the first frame so please generate the prompt to use. The trucks should be moving towards the screen also.
ChatGPT o1’s Response
Begin with a snowy, evergreen forest landscape at twilight. The ground and trees are coated in fresh white powder, and the soft glow of a distant winter sunset casts a warm, golden light over the scene. In the foreground, a classic red holiday-themed Coca-Cola truck, wrapped in festive string lights and bearing the iconic Coca-Cola logo, emerges from the background. Behind it, another identical holiday truck follows closely. Both trucks drive steadily forward along a gently curving, snow-packed road, heading directly toward the viewer’s perspective. Snowflakes drift lazily through the air, and the headlights and vibrant decorations glisten against the quiet, frosty atmosphere. The camera should smoothly track the trucks as they move closer, highlighting the shimmering lights, the deep red paintwork, and the sense of seasonal magic they carry.
I did put this into Sora and it created the video but I wasn’t able to play it due to copyright issues.
It makes sense that it wouldn’t allow me to use the Coca-Cola brand.
But I’m a little disappointed ChatGPT(o1 Model) couldn’t have adjusted the prompt — or warned me about this since I said I was going to use the prompt to make an AI-generated video.
So here is my adjusted prompt:
Begin with a snowy, evergreen forest landscape at twilight. The ground and trees are coated in fresh white powder, and the soft glow of a distant winter sunset casts a warm, golden light over the scene. In the foreground, a classic red holiday-themed truck, wrapped in festive string lights and bearing a generic soda logo in flowing white script, emerges from the background. Behind it, another identical holiday truck follows closely. Both trucks drive steadily forward along a gently curving, snow-packed road, heading directly toward the viewer’s perspective. Snowflakes drift lazily through the air, and the headlights and vibrant decorations glisten against the quiet, frosty atmosphere. The camera should smoothly track the trucks as they move closer, highlighting the shimmering lights, the deep red paintwork, and the sense of seasonal magic they carry.
I think the scene looks great except that Sora didn’t completely follow directions.
The video of the trucks showed them head-on coming toward the screen. I didn’t see a generic soda logo either.
Next Scene — Polar Bear Emerging out of the Water
To attempt to get Sora to follow my directions more closely — I pasted two screenshots into ChatGPT. I had a starting picture of a polar bear underwater, and a second ending screenshot of the polar bear putting its head outside the water and looking out at the icy world.
ChatGPT-o1 Response
Begin beneath the surface of a tranquil, icy sea beneath a starry polar night sky. The water is crystal-clear, allowing the camera to capture the gentle sway of floating ice and shimmering particles of frost. A graceful polar bear drifts into view, its white fur softened by the cool blue light filtering down from above. Its eyes open thoughtfully as it kicks its powerful hind legs and paddles its front paws, slowly ascending toward the water’s surface.
As the bear emerges, the camera follows closely, passing through the rippling surface into the crisp Arctic air. Snowflakes dance lazily overhead, and the soft glow of the aurora shimmers across distant ice floes. The polar bear’s fur sheds droplets of water, and its breath steams in the cold, reflecting the quiet grandeur of the wintery landscape. Finally, the camera lingers on the bear’s face, capturing its calm, inquisitive expression as it lifts its head and gazes out at the serene, icebound world
Sora’s Polar Bear Video
Again, the video looks great, but Sora didn’t follow directions very well.
I was clear that the polar bear should start under-water and then emerge out of it in the prompt
“As the bear emerges, the camera follows closely, passing through the rippling surface into the crisp Arctic air.”
The polar bear in the video I generated with Sora — didn’t come near to resurfacing to show any scenery.
These were my first attempts at making videos with Sora.
The videos look great, but I will need to find a better prompting strategy.
Next time, I will start with an image as well — since the video will need to generate the video starting from this keyframe.
My guess is — using less prompting language should help as well.
For AI Image Generation, often detailed prompts generated by ChatGPT help to make great AI images.
But in the case for AI generated videos — maybe less is more.
Note:
I just started an AI Image generation platform Cartario — Train a custom AI model on your own photos or product photos to create the ultimate brand.
Try out Cartario for free (we just launched this and people are loving it!)
You can check out our AI podcast here.
My other Channels:
👉 Sign up to our free 5-Day email course to grow 🚀 and earn💲in the AI age
You can also sign up for my newsletter on how to use AI to earn more money.
Check out our YouTube Channel
Follow us at our website: AI Growth Guys