tags:
- text-to-image
- stable-diffusion
- lora
- diffusers
- template:sd-lora
widget:
- text: >-
RAW photo, male focus, candid, medium shot, low angle, Pno, as Alexander
Hamilton, Hamilton An American Musical, indoors, on stage, theatre
<lora:XenoDetailer_v2:0.6> <lora:Pno_v1-gf:0.9>
parameters:
negative_prompt: >-
(((girl))),(((boy))), (((underage))), (((watermark))), signature, low
contrast, underexposed, overexposed
output:
url: images/00128-3844552887.png
- text: >-
RAW photo, male focus, candid, medium-wide shot, Pno, sitting, typing,
college, student, masterpiece, computer lab, earbuds
<lora:XenoDetailer_v2:0.6> <lora:Pno_v1-gf:0.9>
parameters:
negative_prompt: >-
(((girl))),(((boy))), (((underage))), (((watermark))), signature, low
contrast, underexposed, overexposed
output:
url: images/00095-1810333198.png
- text: >-
RAW photo, male focus, candid, medium-wide shot, Pno, vibrant colors,
masterpiece, beach, sunset <lora:XenoDetailer_v2:0.6>
<lora:Pno_v1-gf:0.9>
parameters:
negative_prompt: >-
(((girl))),(((boy))), (((underage))), (((watermark))), signature, low
contrast, underexposed, overexposed
output:
url: images/00080-2283299861.png
- text: >-
RAW photo, male focus, portrait, close-up shot, from front, Pno, shot on
Hasselblad, professional headshot, muted colors, masterpiece, ruggedly
handsome <lora:XenoDetailer_v2:0.7> <lora:Pno_v1-gf:1>
parameters:
negative_prompt: >-
(((girl))),(((boy))), (((underage))), (((watermark))), signature, low
contrast, underexposed, overexposed
output:
url: images/00054-1335943862.png
base_model: stablediffusionapi/v1-5-pruned-v6
instance_prompt: Pno
license: cc-by-nc-nd-4.0
Pino Agbayani (Ottoverse original character)
- Prompt
- RAW photo, male focus, candid, medium shot, low angle, Pno, as Alexander Hamilton, Hamilton An American Musical, indoors, on stage, theatre <lora:XenoDetailer_v2:0.6> <lora:Pno_v1-gf:0.9>
- Negative Prompt
- (((girl))),(((boy))), (((underage))), (((watermark))), signature, low contrast, underexposed, overexposed
- Prompt
- RAW photo, male focus, candid, medium-wide shot, Pno, sitting, typing, college, student, masterpiece, computer lab, earbuds <lora:XenoDetailer_v2:0.6> <lora:Pno_v1-gf:0.9>
- Negative Prompt
- (((girl))),(((boy))), (((underage))), (((watermark))), signature, low contrast, underexposed, overexposed
- Prompt
- RAW photo, male focus, candid, medium-wide shot, Pno, vibrant colors, masterpiece, beach, sunset <lora:XenoDetailer_v2:0.6> <lora:Pno_v1-gf:0.9>
- Negative Prompt
- (((girl))),(((boy))), (((underage))), (((watermark))), signature, low contrast, underexposed, overexposed
- Prompt
- RAW photo, male focus, portrait, close-up shot, from front, Pno, shot on Hasselblad, professional headshot, muted colors, masterpiece, ruggedly handsome <lora:XenoDetailer_v2:0.7> <lora:Pno_v1-gf:1>
- Negative Prompt
- (((girl))),(((boy))), (((underage))), (((watermark))), signature, low contrast, underexposed, overexposed
Model description
From his character bio: "People have often underestimated Pino's abilities, but his journey is a testament to the power of perseverance and self-discovery."
OK, so: this is my first male character. Also, I used a different wildcard prompt to make the training set, so unlike the others, I didn't have the original prompts to make the training captions. Not a problem, I'll just fire up WD1.4 tagger then take a manual pass at the results, right? Trained this model and...did not expect the results I got.
My best guess is instead of "solo focus" on the female models, I had "male focus" in the captions. And the optimizer took that as, maleness is a thing I shoud ignore. So despite the facial hair and some muscular shots, the images this model generates are a little unpredictable.
Then I found out if I added back the "male focus" or "man" tags to the prompt, I got what I expected, To experiment, I added "female focus" and found it pretty reiiably gives me a woman. Maybe a fun model to toy with to do gender swap stories, or if you take the details above, maybe to see if you can replicate it.
For now, I'm tagging this v1.0-gf, for "gender-fluid", since I don't have time right now to try to correct it (if I should correct it at all).
Trigger words
You should use Pno to trigger the image generation.
Download model
Weights for this model are available in Safetensors format.
Download them in the Files & versions tab.