Downloads · 30 days
0
rehan-fal/flux-2-klein-4b-paper-diorama-lora
flux-2-klein-4b-paper-diorama-lora is a text-to-image model from rehan-fal. Use it when you need an image from a text prompt. The card lists the license as apache-2.0.
The reference adapter for example 04.4 "Train a LoRA → serve it": trained by that folder's own trainer app on the 16 klein-generated images in dataset/paper-diorama-16.zip (1800 steps, lr 1e-4, rank 16 — full
Downloads · 30 days
0
Access
Public
Updated Aug 5, 2026
Repo size
126 MB
Likes
0
Public
Click a slice to open those files.
.safetensors96.4 MB · 77%
From the Hugging Face model README
The reference adapter for example 04.4 "Train a LoRA → serve it":
trained by that folder's own trainer app on the 16 klein-generated images in
dataset/paper-diorama-16.zip (1800 steps, lr 1e-4, rank 16 — full
config in config.json-style echo inside the example).
pprdrma style, — prepend it to your prompt.black-forest-labs/FLUX.2-klein-4B at 4 steps,
lora_scale ≈ 1.0 (trained on klein-base-4B, BFL's recommended split).| base model | lora_scale 0.7 | lora_scale 1.0 |
|---|---|---|
![]() | ![]() | ![]() |