Guides
Step-by-step guides for every FatThumb feature.
Upload any image and include elements from it in your thumbnail — a product, a scene, a graphic — without it overriding the face or style.
Connect your own OpenAI and Gemini API keys to generate thumbnails at near-zero marginal cost — no monthly credit limits, pay only what the model charges.
Generate a 1280×720 PNG thumbnail in under 60 seconds — pick a lane, describe what you want, and let the AI handle the rest.
Concrete prompting strategies for the Quick and Advanced lanes — what to include, what to skip, and how the 4-block composition scaffold uses your words.
Paste a YouTube URL or upload an image to borrow its visual style — colors, composition, mood — without copying the face.
Upload 1–5 face photos once. Every thumbnail you generate after that reuses your exact face — no retouching required.
Podcast Mode in the Quick lane lets you lock the host's face and swap in a guest per episode — same style, different guest, every time.
Quick is template-first and fast. Advanced is prompt-first and flexible. Here's how to pick the right lane for each video.
Understand the difference between a generation and a thumbnail, how to browse your history, and what you can do with each output.