What the eyes are on
Name what she looks at — does the object appear too?
The previous comparison was direction; this one is object. Name the target and the model draws it in — without the object appearing anywhere else in the prompt.
Every cut below shares one seed, one base prompt and one sampler setting, and differs only by the tag marked in amber.
What each cut shows
looking at phone- A hand and a phone appear together and the gaze drops. One tag brought a prop and a pose at once.
looking at mirror- It builds the mirror and her reflected front view, while she herself turns her back — the biggest addition of the four.
looking over eyewear- Sunglasses slide down the nose and the eyes come over the top — eyewear that was never asked for.
looking to the side- The one without an object: direction moves, nothing is added.
How these were made
Shared prompt
masterpiece, best quality, amazing quality, very awa, 1girl, solo, mature female, adult woman, short black hair, brown eyes, white blouse, upper body, plain grey background, soft lighting, <tag slot>
Tap a tag to carry it over to the builder
Negative prompt
nsfw, nude, topless, bikini, swimsuit, lingerie, underwear, cleavage, suggestive, school uniform, pleated skirt, child, loli, teenage, bad quality, worst quality, worst detail, censor, jpeg artifacts, watermark, signature, text, extra limbs, bad hands, bad anatomy
Model
rinFlanimeIllustrious_v30 · euler_ancestral/normal · steps 28 · CFG 5 · 896×1152 · seed 777030
Takeaway
An object-directed gaze tag doubles as a prop tag. If you do not want the object, use a direction tag instead.
How far this result reaches
What the model adds is outside your control — you do not get to pick the phone or place the mirror.
Measured on a single Stable Diffusion family checkpoint. The same tags can behave differently on NovelAI, Midjourney or another checkpoint. This is a reference point, not a verdict.