valkyriesorder3flux-klein

<p align="center"><a href="https://noema.art"><img src="https://huggingface.co/noema-art/noema-brand/resolve/main/noema-banner.png" alt="NOEMA — run this model privately at noema.art" width="100%"></a></p>

valkyriesorder3flux-klein

NOEMA — privacy-by-construction generative studio.
Run this model privately at noema.art · no email · pay anonymously.

FLUX.2 [klein] 4B conversion of noema-art/valkyriesorder3flux (originally FLUX.1-dev). Trigger: "valkyriesorder".

Trigger word: valkyriesorder Retrained onto FLUX.2 [klein] 4B from noema-art/valkyriesorder3flux (FLUX.1-dev).

Sample Outputs

Usage (Diffusers)

import torch
from diffusers import Flux2Pipeline

pipe = Flux2Pipeline.from_pretrained(
    "black-forest-labs/FLUX.2-klein-4B", torch_dtype=torch.bfloat16
).to("cuda")
pipe.load_lora_weights("noema-art/valkyriesorder3flux-klein")
image = pipe("valkyriesorder, a character portrait", guidance_scale=4.0, num_inference_steps=25).images[0]
image.save("out.png")

Recommended Settings

Training Details

• Base: black-forest-labs/FLUX.2-klein-base-4B
• Steps: 4000 · Network: LoRA rank 32 / alpha 32
• Optimizer: adamw8bit, lr 1e-4 · Scheduler: flowmatch
• Resolution: 512, 768, 1024 (multi-res bucketed) · Precision: bf16 train / fp16 save

Reproduction

This repo includes the full dataset/ (20 image-caption pairs) and the exact config.yaml so the LoRA can be retrained as-is.

About

NOEMA is a privacy-by-construction generative studio. Run this model — and the rest of the catalogue — privately at noema.art: no email to start, pay anonymously, go fully private anytime.

_{NOEMA · a complete studio, completely private · <a href="https://noema.art">noema.art</a>}