For best outputs:

Version 2.0:

Sampler: DPM++ 3M SDE Karras.

Steps: 26, cfg 7.5, CLIP skip 2.

VAE: none required.

Pos prompt: quality keywords of your please may help overall detail, like for example,

"(masterpiece), top quality, best quality, official art," ect.

Neg prompt: "(worst quality, low quality, normal quality:2), lowres, split_screen, framed border, childlike paintings, sketches, bad anatomy, (monochrome), (grayscale), FastNegativeV2, badhandv4", being the embeddings "FastNegativeV2" and "badhandv4 included in my v20 example runs.

Version 1.0:

Sampler: DPM++ 2M Karras.

Steps: 26, cfg 7.5, CLIP skip 2.

VAE: blessed2_fp16.safetensors.

Pos prompt: add the LoRa "add_detail" with a weight of 0.7 at the end of your pos (included in my v1.0 example runs).

Neg prompt: "BadDream, negative_hand-neg, (worst quality:2), (low quality:2), (normal quality:2), lowres, bad anatomy, (monochrome), (grayscale), text, signature, watermark, logo", being the embeddings "BadDream" and "negative_hand-neg included in my v1.0 example runs.

What changes from v1.0 to v2.0: input_image feature, overall detail and shadows improvements.

Checkpoint Merged.