Prompt guide · AI image
An editorial close-up built around a curved hat shadow, red lips and a solid red background.
An editorial close-up built around a curved hat shadow, red lips and a solid red background. The curved hat shadow divides the face into two tonal zones. Red lips connect the illuminated lower face to the red background, while the black clothing and hat should remain separate shapes rather than merging into one dark mass.
Reuse the complete prompt with Nano Banana Pro, keeping the broad black fedora, black turtleneck and solid red backdrop. Cast the hat shadow across the upper face while leaving the eyes readable. Adjust light direction or background colour, then check the eyes, lips and separation between hat and backdrop. Limitations: The archive is 833 × 1199 pixels; the prompt fixes no ratio. The proposed Studio 2:3 is an approximate starting format and must not crop the source media. The eyes remain visible in the example despite the strongly worded deep-shadow instruction. This intensity variation is disclosed, not described as exact reproduction. Check whether the eyes are still readable inside the shadow and whether the brim retains its curve. Keep the shadow direction consistent on the nose and cheeks when adjusting the light.
Read the complete prompt alongside the result. Identify the subject, framing, lighting and visual style before changing the brief.
Open the selected recipe in Studio with its model and starting settings. Check the available version, frame and resolution before generating; a size written in the source prompt is not an output setting or a guarantee.
Use the customization fields below to rewrite the brief: Subject and clothing / Contrast / Photographic direction. Keep the composition and other constraints that make this example useful. Check the current credit quote before submitting.
Compare the result with the chosen example: subject position, light direction and the balance between foreground and background. Inspect hands, faces, product shapes and any text at full size. Change one variable at a time so you can judge what improved.
This example is attributed to 𝑨𝒇𝒓𝒊𝒏 @afrinxai.
The work remains attributed to its public source; the link preserves access to its original context.
