Negative expressions
How far do anger and sadness separate?
The negative side lands even quieter than the positive one. Only the tags with a physical marker read reliably.
Angry_face did not look angry, so we suspected the suffix and rendered plain angry beside it. The two were nearly identical — this is the model's disposition, not a spelling problem.
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
Angry face- The brow knots and she glares sideways, teeth showing at one corner of the lips — sharp, but short of an outburst.
crying face- Tears are drawn in, making this the only unambiguous cut. Tags with a physical marker read best.
Gloomy face- Gaze down and saturation off — rendered as mood rather than as face.
frustrated face- Comes with the cheek propped on a fist, and the knitted brow and flushed cheek set it apart from
Gloomy.
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, portrait, 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 777022
Takeaway
Do not lean on one tag for a strong emotion. Name the physical markers — tears, a furrowed brow, clenched hands — that can actually be drawn.
How far this result reaches
The restraint is this checkpoint's temperament. The same tags can be far more theatrical elsewhere.
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.