A prompt for generating a hyper-realistic cinematic portrait of a woman imagined as a celestial archivist suspended in zero gravity, featuring translucent robes, holographic glyphs, and soft cosmic lighting.
A JSON prompt designed to generate an image with a Y2K-style smartphone flash photo aesthetic. It emphasizes preserving 100% facial fidelity and realistic skin texture, using direct flash mixed with colored light, and depicting a female blogger archetype
A detailed, multi-step prompt template designed for Nano Banana Pro to generate a clean, modern food recipe infographic based on an uploaded image (in this case, a hamburger). The prompt specifies visual style, layout, required Chinese text, ingredient li
A structured prompt for Gemini Nano Banana Pro to generate an ultra-realistic, cinematic portrait of a fit, handsome man on a city terrace, emphasizing athletic discipline and modern aesthetic. It strictly requires preserving the facial identity from an u
A highly detailed, structured prompt for generating a conceptual photographic portrait using an extreme high-angle (top-down) perspective. It emphasizes strong wide-angle lens distortion, extensive negative space, cinematic studio lighting, and a cool, co
A prompt for creating a hyper-realistic, editorial-style photograph of a young man in high-fashion mountain gear sitting on volcanic terrain near the Cotopaxi volcano in Ecuador. The prompt emphasizes sharp natural lighting, crisp shadows, and a high-alti
A highly detailed JSON prompt for Gemini Nano Banana Pro to create a hyper-realistic, two-panel vertical portrait simulating iPhone photography. It requires exact face matching for two subjects from reference images, specifying identical composition, ligh
An ultra-realistic night photography prompt for Nano Banana Pro, designed to generate a high-contrast, paparazzi-style image of a woman leaning on a luxury sports car using harsh direct camera flash for a cinematic, moody effect.
An AI text to image generator turns a written prompt into a finished picture you can download—no Photoshop skills, no stock subscriptions, no waiting on a designer. Describe the subject, style, lighting, and composition; the text to image engine renders pixels that match your words. ToolFK ships multiple models including Flux, CodeView3, and FLUX.2 Dev so you can pick speed, detail, or experimental looks depending on the job.
This free AI image generator runs in the browser: type a prompt, choose a model and aspect ratio, and save PNG or JPG in seconds. Flux models excel at photoreal scenes and clean typography; CodeView3 suits stylized illustrations; FLUX.2 Dev offers a balance for iterative creative work. When you need motion next, send the result to our image to video tool or build a full scene with the text to video generator. Stuck on wording? Nano Prompts lists tested templates you can paste and tweak.
Yes. Guests have a daily generation limit; registered users receive more quota, and VIP plans add extra runs. Check the counter on the page before you start.
Use Flux for photoreal output and sharp detail. CodeView3 works well for stylized or illustrative looks. FLUX.2 Dev is a solid default when you want fast iterations without sacrificing too much quality.
Name the subject, art style, camera or lens feel, lighting, and background. Separate ideas with commas or short sentences. Avoid contradictory instructions in one prompt.
You are responsible for rights in your prompt and any trademarks or likenesses you describe. Review ToolFK terms and local laws before using images in paid campaigns.
Most jobs download as PNG or JPG depending on the option you pick. Resolution follows the aspect ratio and model tier selected on the page.
Generate a still here, then open the image to video tool with the same file to add motion. For scenes built purely from words, use text to video instead.
Last updated: July 1, 2026
![TRY IT NOW](/tools/online-text-to-img.html?prompt={
"title": "Hyper-realistic two-panel iPhone portrait — Café (Subject1 + Subject2)",
"mode": "text_to_image",
"format": "two_panel_vertical",
"output_size": "2048x3072",
"references": {
"subject1": "subject1_ref.jpg",
"subject2": "subject2_ref.jpg"
},
"prompt": "Hyper-realistic, ultra-photographic two-panel iPhone portrait of two real people, exact-face-matching to the provided reference images (no facial alterations). Both panels share identical angle, spacing, lens perspective and composition so they read as matched diptychs taken in the same seat at the same lush café. Environment: dense bamboo and tropical foliage around a warm wooden table, warm diffused daylight creating gentle highlights and soft, natural shadows across skin, glass, plants and wood. Upper panel — Subject 1: seated, holding a reflective silver phone with a jeweled ring grip catching small bursts of light; deep-red manicured nails; blue-and-white gingham off-shoulder dress with subtle seersucker texture; white lace bandana with delicate perforations; holding a chilled matcha glass with visible condensation droplets, faint fingerprint smudges on the glass, and swirling gradients of matcha and foam at the rim. Lower panel — Subject 2: seated, mirroring the same candid pose, holding a matte metallic phone with tiny scratches and soft reflections; wearing a cream linen shirt with natural slub texture, brown trucker cap (mesh highlights), and beaded accessories; raising a tall iced latte showing layered tones of milk and coffee, melting ice cubes, and misty condensation. Across both panels: micro-details — water rings on the table, smudges on the glass barrier, uneven leaf coloration in foliage, subtle lens glare, micro-reflections on metal and glass, tiny fabric frays and natural skin pores; maintain lifelike texture and micro-imperfections to reinforce authenticity. Lighting: soft, warm, diffused daylight, beautiful skin tones, delicate specular highlights on glass and metal, accurate subsurface scattering on skin. Camera: iPhone portrait-style framing, shallow depth of field with natural bokeh, crisp focal plane on faces and held glasses, matched focal length and angle between panels. Color/tonal: natural, filmic dynamic range, no oversaturation; preserve subtle tonal shifts of matcha and coffee. Composition: two vertically stacked panels, matching framing, consistent perspective and scale so both subjects read as identical-session captures. Strict requirement: faces must match the provided reference images 100% — do not alter facial features, expressions, or identity. Output should be photorealistic and indistinguishable from high-end phone photography.",
"negative_prompt": "cartoon, painting, CGI, 3D render, anime, stylized, over-saturated, over-contrast, extra limbs, mutated, deformed, blurred, lowres, "){kind=link}