Show HN: I built an AI tool that generates click-worthy YouTube thumbnails

Show HN: I built an AI tool that generates click-worthy YouTube thumbnails(thumbnailx.com)

5 points by gits997 1 year ago | 4 comments

Hey HN! Software engineer here. As an amateur youtuber, I've been struggling with thumbnails, especially since I don't have any design or editing skills.

Decided to give it a try and make my own automatic thumbnail generator. Me being an ADHD person, made sure that there's no parameter tuning and shape choosing and fuss like that, just simply give prompt and generate thumbnail.

Increased my CTR from 1.2% to 2.3% (faceless fantasy books niche on youtube)

Thumbnail X (completely free for now) - https://thumbnailx.com/

Just wanna add that this is by no means a perfect thumbnail maker for huge YouTube moguls with millions of subscribers but I'm sure it's gonna be extremely helpful for beginner youtubers and medium size channels struggling with thumbnails.

Tools used: 1. Cursor for development (shoutout to claude oppus 4.0) 2. Few image generators (primarily ideogram, but leonardo ai, deep ai and gpt 4o as fallbacks) 3. Few llms (chatgpt, claude, gemini for validating images and prompt enhancement) 4. AWS and Redis for storage & caching 5. Digital ocean for hosting and db 6. Python

nbbaier 1 year ago |

The example you give is very clearly AI generated. Has there been any work done on how different such thumbnails behave vs. clearly non-AI thumbnails?

gits997 1 year ago | |

Thanks for the comment. Appreciate the interest.

I did the experiment for my fantasy youtube channel (https://www.youtube.com/@AIBookBytes) and CTR increased from 1.2% to 2% which for me was a huge deal.

This is probably not the tool for super high subscriber youtube channels but from amateurs to mid size youtube channels could be very helpful. Obviousy it is clear that it's AI generated by many channels use that and still are successful. As long as expressions are clear and text is bold and visible, should be good.

eth_hack77 1 year ago |

Sounds interesting! Do you train your image model / customized existing one or what is the source for the thumbnail images?

gits997 1 year ago | |

No, I use existing model(s) as a base along with LLMs to: 1. enhance the prompt and add technical details that are important for high quality thumbnails (e.g. high contrast image) 2. validate the generated images and choose the best one.