<<<V1.0 Version>>>

Sample images are all generated locally with 4090 (N), and the effects differ from online generation!

Recommended large models: Juggernaut XL or other realistic SDXL models

Recommended LoRa weights: 0.6 0.8 (recommended 0.8)

Trigger words: None

Recommended prompts:

nthropomorphized animal<anthropomorphized animal> Pet styling description (e.g.: a dog is wearing medieval Templar armor...) Scene description (e.g.: historical portrait<historical portrait>, dignified expression<dignified expression>...) Additional tags (e.g.: 16th century fashion<16th century fashion>, warm color palette<warm color palette>, soft lighting<soft lighting>...)... Feel free to add elements or use your treasured prompts~

Its characteristic is the ability to generate realistic images of medieval-themed decorative cats and dogs!

Negative prompts: (Write them as needed)

(worst quality, low quality, normal quality:2), Text, Watermark, logo,

Sampling method: Restart (No specific requirement, up to you)

Iteration steps: 30 (Up to you)

Recommended dimensions: 1:1/2:3/3:4 (For SDXL, just run 1024×1024 or 768×1024 directly, for high resolution use 2x upscale)

Prompt guidance coefficient (CFG): 7.0

High resolution repair: 4x UltraSharp. If an image meets the requirements, use high-resolution upscale, with redraw magnitude of 0.3 0.4, or directly redraw with magnitude above 0.5 for upscaling~

<Notes>

Some natural scene prompts may have less noticeable effects, and subsequent versions will enhance these scene characteristics;

Since the LoRA model is trained based on the SDXL_base base model, using different large models may yield varying results;