Views
No views yet
ohisee
Retrained onto FLUX.2 [klein] 4B from ms2stationthis/ohiseeflux (FLUX.1-dev). |  |
|---|---|
| ohisee, This is a digital drawing in a soft, pastel style, featuring a young anime-style… | ohisee, This is a digital drawing in a soft, pastel-like ohisee style, featuring a young… |
 |  |
| :---: | :---: |
| ohisee, mystical aesthetic. This "ohisee" style art pairs well with this digital drawing'… | ohisee, This is a digital drawing in a semi-realistic, anime-inspired style, manifesting… |
1import torch
2from diffusers import Flux2Pipeline
3
4pipe = Flux2Pipeline.from_pretrained(
5 "black-forest-labs/FLUX.2-klein-4B", torch_dtype=torch.bfloat16
6).to("cuda")
7pipe.load_lora_weights("ms2stationthis/ohiseeflux-klein")
8image = pipe("ohisee, a character portrait", guidance_scale=4.0, num_inference_steps=25).images[0]
9image.save("out.png")| LoRA strength | Guidance | Steps | Resolution |
|---|---|---|---|
| 0.8–1.0 | 4.0 | 25 | 1024×1024 |
dataset/ (21 image-caption pairs) and the exact config.yaml so the LoRA can be retrained as-is.