Prompt: Generate a Video Thumbnail

Generates one high-contrast video thumbnail with a clear focal point and space reserved for text, sized to read at small preview size.

The prompt

Open in Studio

Generate ONE video thumbnail image with the following exact specification.

Subject: {{SUBJECT — e.g. a person looking shocked while pointing at a laptop screen}}.

Topic of the video: {{VIDEO TOPIC — e.g. a coding mistake that cost a company thousands}}.

Composition: subject fills {{APPROXIMATE FRAME PERCENTAGE — e.g. 60%}} of the frame, positioned {{SUBJECT POSITION — e.g. right third of the frame}}, with the {{OPPOSITE THIRD — e.g. left third}} left visually clear for a text overlay to be added afterwards.

Style: {{STYLE — e.g. bold, high-contrast, slightly exaggerated expression, saturated colours}}, aspect ratio 16:9.

Constraints:
- Exactly one clear focal point (the subject); no secondary figures or distracting background elements competing for attention
- High contrast between the subject and the background so the image stays legible when scaled down to a small preview size
- Do not render any text, numbers or logos anywhere in the image; any text will be added separately afterwards
- No blur across the subject; the subject must be in sharp focus even if the background is softly blurred

Example: a shocked expression with an arm pointing off-frame reads clearly even at thumbnail size; a wide, busy scene does not.

Return only the generated thumbnail image, with no accompanying text or caption.

Before generating, check that the subject position and the clear third don't overlap, and that the style requested wouldn't obscure the single focal point.

Example

Input

SUBJECT — e.g. a person looking shocked while pointing at a laptop screen: a person looking shocked while pointing at a laptop screen
VIDEO TOPIC — e.g. a coding mistake that cost a company thousands: a coding mistake that cost a company thousands
APPROXIMATE FRAME PERCENTAGE — e.g. 60%: 60%
SUBJECT POSITION — e.g. right third of the frame: right third of the frame
OPPOSITE THIRD — e.g. left third: left third
STYLE — e.g. bold, high-contrast, slightly exaggerated expression, saturated colours: bold, high-contrast, slightly exaggerated expression, saturated colours

Output (a written description, not a generated image)

What a good result should look like: A bold 16:9 thumbnail showing a person with a wide-eyed, exaggerated shocked expression, positioned in the right third of the frame and filling roughly 60% of it, one arm raised and pointing toward a laptop screen just in frame beside them. The subject is lit with strong, saturated colour and sits in sharp focus against a softly blurred, dark background that throws them into high contrast, keeping the image legible even at a small preview size. The left third of the frame is visually clear and uncluttered, deliberately left open for a text overlay to be added afterwards. There is exactly one focal point — no second person or busy background element competes for attention. No text, numbers or logos appear anywhere in the image itself.

Why it works

Splitting the frame into a subject third and a clear third solves the real thumbnail problem of leaving room for text without it fighting the image.
Banning rendered text keeps the image as a clean base layer, since thumbnail text is normally added separately and needs to stay editable.
The pre-generation check for overlap between the subject and clear third catches a composition that would leave no room for the text overlay.

Why the basic version fails

“Make a thumbnail for my video about a coding mistake” produces a busy, centred image with no space reserved for a headline, or one where the subject and background compete for attention — both routinely unreadable once shrunk down to the small size thumbnails actually appear at.

What this prompt fixes

Fixing the subject’s position and reserving an explicit clear third is what makes room for a text overlay without a second editing pass to awkwardly crop or recompose the image. Banning rendered text keeps the base image reusable, since thumbnail text is usually swapped or A/B tested separately from the image itself, and the high-contrast requirement is what actually determines whether the thumbnail reads at the tiny size it will really be viewed at.

Tips per model

Gemini responds well to an explicit frame-percentage and position instruction; a vague “subject on one side” request produces inconsistent framing.
ChatGPT’s image generation reliably avoids rendering text when told not to, but will sometimes add a logo-like graphic unless explicitly banned as done here.
Preview the result at actual thumbnail size (roughly 120 pixels wide) before finalising — details that look fine full-size often disappear at real preview scale.