
Photography-level output~Enhanced realism
Based on the model fine-tuned under the Flux architecture, it features extremely high aesthetic quality fashion photography images and classical style photography images for training and optimization, with significant improvements in composition aesthetics and output details. Compared to the original version of Flux, the skin texture is more delicate and realistic.
Sentence recommendation: Natural language
Sampler recommendation: Euler, other samplers are not as effective
Positive prompt recommendations:(masterpiece:1.2), best quality, high resolution, unity 8k wallpaper, (illustration:1), perfect lighting, extremely detailed CG, finely detail, extremely detailed, soft lighting and shadow, soft yet striking lighting, film grain:1.2, (skin pores:1.2), (detailed skin texture:1), ((solo:1.5)), Detailed face, (see through:1.1), misty, Low Key:1.1, depth of field,
Steps: Recommended 20, the larger the steps the better, but after 20 the improvement is minimal
Size: 768, 1536, enlarge by 1.5 to 2 times,
CFG: Recommended ~3.5
Basic issues: Too much noise~Increase CGF, unclear image quality~Increase iteration steps, deformed hands and feet~Increase iteration steps
Photography-level output~Enhanced realism
Based on the model fine-tuned under the Flux architecture, it features extremely high aesthetic quality fashion photography images and classical style photography images for training and optimization, with significant improvements in composition aesthetics and output details. Compared to the original version of Flux, the skin texture is more delicate and realistic.
Sentence recommendation: Natural language
Sampler recommendation: Euler, other samplers are not as effective
Positive prompt recommendations:(masterpiece:1.2), best quality, high resolution, unity 8k wallpaper, (illustration:1), perfect lighting, extremely detailed CG, finely detail, extremely detailed, soft lighting and shadow, soft yet striking lighting, film grain:1.2, (skin pores:1.2), (detailed skin texture:1), ((solo:1.5)), Detailed face, (see through:1.1), misty, Low Key:1.1, depth of field,
Steps: Recommended 20, the larger the steps the better, but after 20 the improvement is minimal
Size: 768, 1536, enlarge by 1.5 to 2 times,
CFG: Recommended ~3.5
Basic issues: Too much noise~Increase CGF, unclear image quality~Increase iteration steps, deformed hands and feet~Increase iteration steps