Today we are excited to announce the arrival of text-to-video prompts to our marketplace, and the addition of two new models: Sora, and KLING AI.

PromptBase started as a marketplace for GPT-3 prompts, then added text-to-image prompts with the arrival of DALL-E and Midjourney. As generative AI has progressed, we’re thrilled to now also add these state-of-the-art video models.

This short blog post is aimed at creators on our platform to discuss the nuances of selling text-to-video prompts. If you’re looking to explore video prompts that have already been submitted, click the button below.

Selling your first text-to-video prompt

Submitting your first text-to-video prompt is simple. If you’ve already submitted an image prompt to PromptBase, it’s not too dissimilar.

The first thing you’ll need to provide is your prompt. As with all prompts on PromptBase, this will need to be a template with variables in square brackets. For example:

A cinematic instagram reel of a [food item] exploding in mid air on a black photo studio background.

For guidance on what prompts can be submitted to PromptBase, please refer to our prompt guidelines.

When submitting your prompt, we require 4 example videos to be submitted along with it, as well as the variables used to create them. Example videos can be up to 25MB in size.

As a bonus, we’ll take these 4 videos and generate a 2x2 video collage of all your example videos to really showcase your work.

Thumbnails

With image prompts, you can customize the thumbnail of your prompt using any of the example images you submitted with it.

With video prompts, you can do the same, except that you can choose any frame from any of your videos to use within the thumbnail.

We have updated the thumbnail editor to add a new timeline scrubber so that you can navigate through your videos to select the perfect shot to make your thumbnail stand out.

App builder + generations

As with all other prompts on our site, they are compatible with the PromptBase app builder, and can be combined with other prompts and blocks to build complex AI apps.

If a prompt you made is used within the app builder by another user, then 50% of our pro-tier yearly plan credit cost (currently: $0.0020) multiplied by the number of credits it cost to run the prompt are paid out per valid prompt use (for more info on app monetization, check out our guide here).

For video prompts, due to the high computational costs, credit costs are much higher than image or text prompts, meaning you could earn much higher revenue per generation.

Note: Generating videos on PromptBase via text-to-video prompts is not currently live, but will be rolling out fully over the next few days.

Conclusion

Text-to-video prompts present an exciting opportunity for creators on our platform. Right now, generating videos via AI is expensive compared to other mediums (text or images); meaning that having a quality prompt becomes much more important; especially if this prompt is used in vital workflows.

We’re super excited to see what you create!

If you have any feedback, questions or suggestions about this feature, please don't hesitate to reach out via the email on our contact page, or send a message to @ben.