Entry for https://civitai.com/bounties/1060

This is a LoRA for creating beefcake women. Strong enough to carry a pot of soup for the whole family and then some.

Base models are biased towards beefcake men. From my testing this LoRA makes prompting generate females more regularly - without it, sometimes the base model would generate a male form even with "woman" in the prompt.

Most weights work well if you want to use this as a "slider" - see example images.

Low weight 0.2-0.3: will bias generations towards the correct gender whilre retaining the most of the base model's muscle definition.

Weight 0.8-1.0: "Full weight" for this LoRA. Increase the weight on this LoRA from the starting range of 0.2-0.3 until the desired muscularity is reached. Above this may cause artifacting.

No specific trigger word, just describe the model as you would without a LoRA: muscular, ripped, flexing, buff, bodybuilder, fitness model, etc

Prioritized stability with other likeness/LoRA during development/testing

Some base models do not need this LoRA as they have enough information in the training data. If you are satisfied with the performance from your base model, you do not need to use this LoRA in addition.