Views
No views yet
petravoice
Retrained onto FLUX.2 [klein] 4B from ms2stationthis/petravoiceflux2 (FLUX.1-dev). |  |
|---|---|
| petravoice, This petravoice image is a digital artwork featuring a stylized, abstract por… | petravoice, This is a digital artwork featuring a stylized, anime-inspired portrait of a… |
 |  |
| :---: | :---: |
| petravoice, This petravoice image is a digital illustration in a stylized, anime-inspired… | petravoice, This petravoice image is a digital illustration with a soft, ethereal quality… |
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/petravoiceflux2-klein")
8image = pipe("petravoice, 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/ (22 image-caption pairs) and the exact config.yaml so the LoRA can be retrained as-is.