If you landed here after seeing a girl-in-suit portrait on Instagram and wanting the exact prompt behind it, this is the page. Salwar suit prompts are harder than saree prompts for one simple reason: the word "suit" on its own makes most AI models draw a business blazer. You have to name the garment, the fabric and the dupatta, or the result is a woman in office wear standing in a field.
Every prompt below names all three. They cover the regional looks people actually search for (Punjabi phulkari, Lucknowi chikankari, Patiala) plus a velvet winter suit and an everyday cotton suit. Each ends with a face-lock instruction, so the woman in the output still looks like you.
How to use these salwar suit prompts
Upload one clear, front-facing photo first, then paste the prompt. They work as written in Gemini and ChatGPT.
- Use a sharp, well-lit reference. A blurry or heavily filtered source photo is the main reason the face drifts.
- Keep the garment words. "Salwar kameez", "Patiala salwar" and "dupatta" are what stop the model from drawing a western suit.
- Swap colours freely. Change the magenta to mint or the cream to black; leave the fabric and drape descriptions alone.
- Ask for 4:5. Add "4:5 portrait" to the end of any prompt so the image fills the Instagram feed.
Classic regional salwar suit prompts
Using the uploaded photo as a character reference, recreate the woman standing in a blooming mustard field in Punjab in the late afternoon, shot at eye level with a 50mm lens, wearing a plain cream cotton salwar kameez with a hand-embroidered phulkari dupatta in orange, pink and green draped over one shoulder and lifted by the breeze, silver jhumkas, a long braid with a colourful tassel. Golden hour backlight with a warm rim on her hair, a soft smile looking just off-camera, rich saturated colours, shallow depth of field, photorealistic, natural skin texture, keep the face 100% identical to the uploaded photo.
The phulkari dupatta carries this whole look, so keep its colour list. Gemini follows "lifted by the breeze" well and adds natural movement.
Using the uploaded photo as a character reference, recreate the woman seated on the carved stone steps of an old Lucknow haveli courtyard, wearing a white chikankari kurta with matching straight pants and a sheer chikankari dupatta, small pearl studs, hair in a loose low bun. Soft diffused morning light falls through an arched doorway onto pale pink walls, hands resting calmly on her knees, a quiet elegant mood, soft film-like tones, photorealistic, keep the face 100% identical to the uploaded photo.
White-on-white needs soft light. If the embroidery looks flat, add "visible fine shadow-work embroidery on the fabric".
Using the uploaded photo as a character reference, recreate the woman sitting on the ledge of a rooftop terrace at sunset, wearing a deep magenta Patiala salwar with a short embroidered kurti, a teal dupatta slipping off one shoulder and gold-toned juttis visible, laughing as she looks down at her shoes. The dusk sky is pink and orange, a city skyline blurred behind her, warm sunset light, candid and unposed, photorealistic, keep the face 100% identical to the uploaded photo.
The pleats of a Patiala are the point. "Deep gathered pleats" keeps the model from drawing a plain straight salwar.
Everyday and seasonal suit prompts
Using the uploaded photo as a character reference, recreate the woman walking along a misty hill-station road in winter, wearing a deep maroon velvet salwar suit with a gold zari neckline and a cream pashmina shawl wrapped over her shoulders, holding a steaming kulhad of chai, her breath visible in the cold air. Soft overcast light, pine trees fading into fog behind, a cosy mood, muted warm tones, photorealistic, keep the face 100% identical to the uploaded photo.
A good pick for December and January traffic. Velvet reads well because the prompt asks for a zari edge, which gives the fabric a visible border.
Using the uploaded photo as a character reference, recreate the woman browsing a colourful bangle stall in an old city market, wearing a pastel mint georgette suit with a mirror-work dupatta, stacks of lac bangles glittering on the stall behind her, shot candidly from the side. Midday sun with warm reflected light, the busy background kept soft with shallow depth of field, a genuine laugh, vibrant colours, photorealistic, keep the face 100% identical to the uploaded photo.
Candid prompts like this one look least like AI. "Shot candidly from the side" is the line that does it.
Using the uploaded photo as a character reference, recreate the woman sitting by a window in a college library with an open notebook, wearing a simple printed cotton straight suit in sky blue with a light dupatta pinned on one side, hair in a loose low bun. Soft window light from the left, a calm studious mood, natural colours, photorealistic, gentle film grain, keep the face 100% identical to the uploaded photo.
The most relatable of the six, and the one most people actually post. Swap the library for a cafe or a bus window and it still works.
Why salwar suit prompts usually go wrong
Four mistakes cause nearly every bad result:
- Writing "suit" and nothing else. The model reads it as a business suit. Say "salwar kameez" or name the type: Patiala, straight-cut, sharara.
- Forgetting the dupatta. Without it the outfit looks like a plain kurta and pants. State how it is draped: over one shoulder, across the chest, or slipping off.
- Not naming the fabric. Cotton, georgette, velvet and chikankari all drape differently. The fabric word changes the folds and the shine.
- Dropping the face-lock line. Without it the model smooths skin and reshapes the jaw. Keep the last sentence of every prompt.
Frequently asked questions
Why does Gemini give me a business suit instead of a salwar suit?
Because "suit" is ambiguous. Use the full term "salwar kameez" or "Punjabi suit" in the first line and describe the dupatta. All six prompts above do this, so they avoid the problem.
Do these work in ChatGPT as well as Gemini?
Yes. Both accept an uploaded reference photo and respond to the same descriptive structure. Gemini tends to hold facial likeness slightly better, while ChatGPT follows embroidery and fabric details more literally.
Can I use these without uploading my own photo?
You can, but the AI will invent a new face. For a result that looks like you, upload a sharp front-facing photo and keep the closing face-lock sentence.
Which of these suit prompts works best for Instagram?
The Phulkari Mustard Field and the Patiala Rooftop Sunset do best, because the strong colours read clearly at thumbnail size. Use the 4:5 ratio so the post fills the feed.
