Photo prompts work when they read like a photographer’s brief: the shot type, the subject and what it is doing, the setting, the light, the lens and the textures you want to see. ChatGPT, Gemini and Midjourney all respond to that vocabulary, and the official examples below from OpenAI and Google follow it, alongside prompts shared by the community.
On this page
The five parts of a photo prompt
Google’s developer guide to prompting its image model puts it simply: “For realistic images, think like a photographer.” Its template for photorealistic scenes is a good skeleton for any tool:
A photorealistic [shot type] of [subject], [action or expression], set in [environment]. The scene is illuminated by [lighting description], creating a [mood] atmosphere. Captured with a [camera/lens details], emphasizing [key textures and details]. The image should be in a [aspect ratio] format.
Read as a checklist, a photo prompt needs:
- The shot. Close-up portrait, wide shot, top-down flatlay, macro.
- The subject and the moment. Who or what, and what is happening: “carefully inspecting a freshly glazed tea bowl” beats “a potter”.
- The place. A named setting gives the model a background to build.
- The light. Golden hour through a window, overcast daylight, a single softbox.
- The camera feel and texture. Lens, depth of field, film grain, skin or material texture, and the aspect ratio.
Photo words that change the result
| What you control | Words the vendors use | Source |
|---|---|---|
| Framing and angle | “wide-angle shot”, “macro shot”, “low-angle perspective”, “Dutch angle” | Google developer guide |
| Lens | “85mm portrait lens”, “shallow depth of field (f/1.8)” | Google developer guide; Google Cloud guide |
| Camera type | “a GoPro”, “a Fujifilm camera”, “a cheap disposable camera” | Google Cloud guide |
| Light | “three-point softbox setup”, “Chiaroscuro lighting with harsh, high contrast” | Google Cloud guide |
| Film and color | “as if on 1980s color film, slightly grainy” | Google Cloud guide |
| Realism | “photorealistic”, “real photograph”, “iPhone photo” | OpenAI prompting guide |
One caution from OpenAI: detailed camera specs “may be interpreted loosely”. Treat a camera model or an f-number as a mood word that pushes the look, not as a setting the model will reproduce exactly.
Official photo prompts to start from
A photorealistic portrait (Google)
Google’s filled-in version of its own template:
A photorealistic close-up portrait of an elderly Japanese ceramicist with deep, sun-etched wrinkles and a warm, knowing smile. He is carefully inspecting a freshly glazed tea bowl. The setting is his rustic, sun-drenched workshop. The scene is illuminated by soft, golden hour light streaming through a window, highlighting the fine texture of the clay. Captured with an 85mm portrait lens, resulting in a soft, blurred background (bokeh). The overall mood is serene and masterful. Vertical portrait orientation.
It is in our library as the photorealistic portrait of a ceramicist.
A candid documentary photo (OpenAI)
OpenAI’s image prompting guide advises prompting “as if a real photo is being captured in the moment” and asking for real texture instead of studio polish:
Create a photorealistic candid photograph of an elderly sailor standing on a small fishing boat.
He has weathered skin with visible wrinkles, pores, and sun texture, and a few faded traditional sailor tattoos on his arms.
He is calmly adjusting a net while his dog sits nearby on the deck. Shot like a 35mm film photograph, medium close-up at eye level, using a 50mm lens.
Soft coastal daylight, shallow depth of field, subtle film grain, natural color balance.
The image should feel honest and unposed, with real skin texture, worn materials, and everyday detail. No glamorization, no heavy retouching.
Listing: candid photorealistic sailor photo. The last line is the useful trick: “No glamorization, no heavy retouching” keeps the image from looking like an advert.
A product photo (Google)
The Gemini API image guide has a template for commercial shots:
A high-resolution, studio-lit product photograph of a [product description] on a [background surface/description]. The lighting is a [lighting setup, e.g., three-point softbox setup] to [lighting purpose]. The camera angle is a [angle type] to showcase [specific feature]. Ultra-realistic, with sharp focus on [key detail]. [Aspect ratio].
Google’s own filled-in example is the matte black mug product photo in our library.
Community photo prompts
These were shared on X for specific looks the official examples do not cover. We quote the prompt text; open each post to see the author’s images.
Café flatlay. Shared by @azed_ai on X on October 5, 2026, as a fill-in template for top-down food shots:
Photorealistic top-down shot of [food] with [drink] in a stylish café setting, cheerful fresh atmosphere, [color1] and [color2] palette, elegant food styling, natural textures, soft daylight, subtle shadows, premium editorial food photography, ARRI Alexa look, crisp details, 8K realism
We explain the slots in a fill-in café flatlay prompt for food photos.
Miniature product between two fingers. Shared by @azed_ai on X for small, ad-style product shots. Replace [PRODUCT]:
A high-resolution advertising photograph of a realistic, miniature [PRODUCT] held delicately between a person's thumb and index finger. clean and white background, studio lighting, soft shadows. The hand is well-groomed, natural skin tone, and positioned to highlight the product's shape and details. The product appears extremely small but hyper-detailed and brand-accurate, centered in the frame with a shallow depth of field. Emulates luxury product photography and minimalist commercial style.
Exploding ingredients. Shared by @icreatelife on X for food and drink packshots. Upload your product photo first:
Photograph this product in a dramatic modern scene accompanied by explosive outward dynamic arrangement of the key ingredients fresh and raw flying around the product signifying its freshness and nutritional value. promo ad shot, without text, product is emphasized, with the key brand colors as background.
Photo prompts in ChatGPT, Gemini and Midjourney
- ChatGPT. OpenAI says any format works, from a short line to labeled sections, “as long as the intent and constraints are clear”. Put “photorealistic” in the prompt. Our ChatGPT image prompts guide has the rest of OpenAI’s rules.
- Gemini. Use Google’s template above and state the aspect ratio. For edits of your own photos, see the Gemini AI photo prompt guide.
- Midjourney. Add
--raw. Midjourney’s Raw documentation says it turns off the model’s automatic styling, and “with simple prompts, you’ll often get more realistic, photo-like images”. Set the frame with--ar; our Midjourney parameters cheat sheet lists the rest.
Fixes when the photo looks fake
| Problem | What to add or change |
|---|---|
| Skin or surfaces look plastic | Ask for real texture (“visible pores”, “worn materials”) and add OpenAI’s “No glamorization, no heavy retouching”. |
| It looks like a stock photo | Describe a moment, not a pose: “calmly adjusting a net”, “caught mid-laugh”. “Candid” and “unposed” help. |
| Wrong framing | Name the shot (“medium close-up at eye level”) and the aspect ratio. |
| Flat light | Name one light source and its direction, such as golden hour through a window. |
| Your own photo changed too much | Write “change only” the part you want edited and “keep everything else the same”, as OpenAI’s guide suggests. |
Upload photos of other people only with their consent. Google’s API documentation says every image its models generate carries a SynthID watermark.
