A LoRA trained to generate doggystyle compositions with the subject facing the camera.
This model was trained on a 61 photos of porn stills, it's a little rigid so I am working on an expanded dataset

Main Trigger Word:ffdoggy (essential for activating the primary positioning logic). While it will produce results without the trigger, usingffdoggy is strongly recommended for consistent composition. Additionally using her hips pressed against her ass helps a lot to keep the man perfectly behind her if the composition is complex.

Core Conceptual Keywords:

• Positioning:facing camera,man standing behind a woman,ass up,on elbows,on hands,breasts pressed down,face down,looking down,man's hand on head

• Action/Detail:bouncing breasts,breast hanging down,hair pulling

• Body/Face:curvy,petite,huge breasts,large breasts,medium breasts,small breasts,closed eyes,open mouth,grimacing,pale skin,dark skin

• Accessories:pierced nipples,fishnet leggings,glasses,necklace,tattoos

  • pierced nipples works inconsistently.

What it's capable of:

• Consistent Framing:

  • Excels at maintaining the "woman facing camera" perspective, keeping the man behind her out of focus or partially obscured.

  • Key positioning terms likeon elbows or on hands work reliably to define the base pose.

• Body Diversity:

  • Trained on a range of body types frompetite tocurvy, with a good amount of data for various breast sizes, fromsmall tohuge.

• Can generate different skin tones effectively usingpale skin anddark skin prompts.

• Facial Expressions:

  • Responds well to facial prompts likeclosed eyes,open mouth (for pleasure or shock), andgrimacing to convey a wide spectrum of reactions but is not limited to those.

Known Quirks:

• It's sometimes to rigid in the positioning making more dynamic posing difficult to achieve

• The exact strength may need to be adjusted (typically 0.7-1.0) to balance the pose with other elements in the prompt.