Fresh out of the oven
Spot a bug? These tools are new, so the odd thing may still wobble. If something looks wrong, it probably is. Tell us and we will sort it..
Fresh out of the oven
Spot a bug? These tools are new, so the odd thing may still wobble. If something looks wrong, it probably is. Tell us and we will sort it..
Fresh out of the oven
Spot a bug? These tools are new, so the odd thing may still wobble. If something looks wrong, it probably is. Tell us and we will sort it..
Pinnora's video thumbnail maker renders 16:9 covers for video listings. You write what the video is about and the one reason someone would tap it, then pick a style and a face. Leave the overlay line blank and it writes a short one from your saved customer pains. Up to 12 per run, each on its own angle.
The tool is open above. This is the short version of what to do with it.
Style covers Bold text, Product hero, Lifestyle, Comparison and Before / after. Face emotion sets that person's expression, or No face.
The subject plus the one reason someone would tap it. Fill for me drafts that line and preselects the dials around it.
Variations runs 1 to 12, and each one is a separate render carrying its own angle rather than the same picture again.
Text overlay caps at 40 characters and is drawn exactly as typed. Leave it empty and a hook pass writes two to five words first.
Short answers to what people ask before they start.
Variations is a ruler that runs 1 to 12 and starts at 3, and the button reads Generate 3 whenever it is above one. Each variation runs on its own and settles its own credits. When the project has research behind it, the batch also hands every run a different pain, brand pillar, hook, tone and composition, so eight comes back as eight angles rather than one picture eight times.
No. Text overlay accepts up to 40 characters and the placeholder asks for 4 to 6 words. Whatever you type is passed to the render as the line to draw exactly, huge, with a heavy stroke so it survives any background. Leave it blank and a hook pass runs first, writing a two to five word overlay plus the subject direction from your brief and your project's pains.
Use this project's details sits in the project menu above the controls and starts On. On means the run reads your brand, research, strategy and reference pack, so the hook pass writes from your real pains and desires and the image prompt carries your product category, which is what stops a makeup product being drawn as skincare. Off drops those four sources, the honest setting on the scratch workspace where there is nothing to read.
Every run comes back at 16:9, the frame YouTube and most video listings use. The tool has no aspect control, so a portrait crop is not something you can pick by accident. One run renders one image, which is why Variations is the multiplier rather than a sample count. Finished images are uploaded into the project you generated them in and stay under the History tab.
The Model row starts on Auto, and Auto for thumbnails resolves to openai/gpt-image-2. Pick another image model from that row and the cost follows what you picked. The Prompt Builder behind Fill for me will write your brief and preselect Style, Face emotion and Variations, but it is blocked from setting Model on purpose, so a switch to a pricier model is always a decision you made.
Generating needs a signed-in account. The run checks your role on the project and writes the result into it, and a new organization starts with 10 welcome credits. Tool pages are public, so a run started here also carries an IP limit of 20 generations per 10 minutes on top of the per-account checks. Reading the page and filling the form need no account.
The rest of the family. Every one is free to try and runs on the same engine.
