No sign-up. No install.
AI image prompt
Describe the subject and the scene. You get a prompt that also carries the procedure for getting the same character back.
This is what comes out
Make a product mockup image of a stainless tumbler**Engine-facing content — copy as one noun-phrase block:**
stainless tumbler product mockup, single tumbler, upright three-quarter view, centered composition, full product visible, clean studio background, soft contact shadow beneath the tumbler, large diffused key light from upper left, gentle fill from front right, controlled edge highlight along the metal body, brushed stainless-steel texture, realistic cylindrical geometry, crisp rim and lid separation, restrained reflections, neutral commercial product photography, uncluttered surface, accurate proportions, sharp focus on the tumbler
**Reader-facing direction — hybrid boundary:**
The block above is the only descriptive material intended for the image generator. The notes below are narrative guidance for the person using it: keep the tumbler as the unchanged visual subject, while the background, lighting, and composition provide the controlled presentation around it. Do not add decorative details unless they are supplied or intentionally chosen, because each added prop changes the product mockup's commercial meaning.
If the product has a supplied brand or character description, insert that exact wording into the engine-facing block and repeat it unchanged in every variant. No such description was provided here.Full instruction | What the engine reads it as | 1 cut |
|---|---|
| Left for the engine to decide | 7 things — put them in the request and they go away |
| Set by the instruction | 2/2 |
| Lines you can drop | 1 line |
| Same subject on a second run | No |
The instruction went in as-is; this is what came back, unedited.
One run. Paste it again and the wording will differ.
This is what Mephify is programmed to produce, so it is AI-authored output, published exactly as it came out.
Make yours
Other media documents: AI music promptAI video prompt
Paste it into whichever generator you use. No tool names are baked in.
We use only the facts you provide — nothing invented, and your text is never stored.
Nothing filled in yet — fill at least one field so the prompt has material.
Ultimate prompt
We turn the choices above into an instruction you can paste into your AI.
The same instruction gives different wording on Gemini. Compare on the example page.
Common questions
- Can I use it without an account?
- You make it right on this page, with no account and no login. Copy the prompt or save it as a file.
- Where do I paste what comes out?
- It is not tied to one generator. No tool name and no tool-specific flag goes into the prompt, so it pastes into whichever image generator you use. Aspect ratio and resolution come out as values and words rather than flags.
- Does what I type stay anywhere?
- Worried the subject you described stays somewhere? Our server keeps no text. Only counts such as how many characters you typed remain, and the text goes to the AI provider once.
- Will the same character come back twice?
- Not from description alone. In the sticker-set run the stand-in held the outline, palette and eye treatment steady but could not promise the same dog twice, and the character-sheet run said as much about three views of one person. Describe the subject in the slots, then keep the first render's seed or a reference image for every later one. Read more
- Is it Midjourney-only?
- No. A flag reads as plain characters outside the tool that owns it, so none go in. In the tumbler run above, the stand-in took aspect ratio and resolution from the prompt and defaulted neither.
- Was this actually rendered?
- No image is shown here. The table above is a stand-in playing that engine, read for what the prompt settles and what it leaves open. The tumbler run guessed seven things, every one of them a product fact the request never gave. If you run it for real, tell us what came out.
More ai image prompt examples
- Create a webtoon character sheet — front, side and back views729 wordsChatGPT · Claude · Gemini · Llama · MidjourneyLog
- Create an emoji-sticker set concept based on my dog799 wordsChatGPT · Claude · Gemini · Llama · MidjourneyLog
From the operator's verification log
What to watch for
The instruction sets a fixed-seed and reference-image procedure for follow-up shots, but no seed exists until the first image is made — so the first render is unanchored by design. Supply the product facts and the guesses disappear.
Where not to use it
A stand-in check, not a rendered mockup. It cannot see your product; describe it or, if your service allows, attach a reference image.