Make images of HannahG, a lora designed to specify a distinct face, while other image features are variable. Recommend LoRas compatible with the SD 1.5 family in the semi-realistic to photorealistic range.
Recommended Generation Data:
NewRealityXL ❗ All-In-One Photographic
Sample Generation Data with Output (thatnks to user mistermcluvin for testing):
<lora:HannahG:1>, (full body shot:0.7), realistic photo of a woman, brown eyes, long hair, (masterpiece:1.0), (best quality:1.0), (intricate details), flash photography
Negative prompt: close up view, bad-hands-5, sitting, extra limb, missing limb, floating limbs, (mutated hands and fingers:1.1), disconnected limbs, blurry, amputation, 3d, cgi, airbrushed, model, render, sketch, cartoon, drawing, anime,
Steps: 50
VAE: sdxl_vae.safetensors
Size: 832x1216
Seed: 1173615971
Model: newrealityxl_20
Version: v1.7.0
Sampler: DPM++ 2M Karras
"HannahG: b8a47f2fbf56"
VAE hash: 235745af8d
CFG scale: 6.5
Model hash: c1adaf9d0c