Downloads · 30 days
3
21% of all-time downloads
cheeseworld1flux-klein
noema-art/cheeseworld1flux-klein
cheeseworld1flux-klein is a text-to-image model from noema-art. Use it when you need an image from a text prompt. It is set up for ai-toolkit. The card lists the license as apache-2.0.
FLUX.2 [klein] 4B conversion of ms2stationthis/cheeseworld1flux (originally FLUX.1-dev). Trigger: "cheeseworld".
All-time downloads
14
Public
Repo size
121 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors92.4 MB · 76%
At a glance
- Task
- Text-to-Image
- Library
- ai-toolkit
- License
- apache-2.0
- Access
- Public
- Created
- Jun 28, 2026
- Updated
- Jun 28, 2026
- SHA
- e1abe8dc
Try a prompt
Base models
Model card
From the Hugging Face model README
cheeseworld1flux-klein
FLUX.2 [klein] 4B conversion of ms2stationthis/cheeseworld1flux (originally FLUX.1-dev). Trigger: "cheeseworld".
Trigger word: cheeseworld
Retrained onto FLUX.2 [klein] 4B from ms2stationthis/cheeseworld1flux (FLUX.1-dev).
Sample Outputs
 |  |
|---|---|
| cheeseworld, This is a highly detailed, digital illustration featuring a surrealistic dep… | cheeseworld, This is a highly detailed CGI artwork featuring an anthropomorphic snow leop… |
 |  |
| :---: | :---: |
| cheeseworld, This is a highly detailed CGI rendering of a cartoonish, anthropomorphic pan… | cheeseworld, This image is a highly detailed, digital artwork depicting a surreal scene s… |
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("ms2stationthis/cheeseworld1flux-klein")
image = pipe("cheeseworld, a character portrait", guidance_scale=4.0, num_inference_steps=25).images[0]
image.save("out.png")
Recommended Settings
| LoRA strength | Guidance | Steps | Resolution |
|---|---|---|---|
| 0.8–1.0 | 4.0 | 25 | 1024×1024 |
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
Trained on StationThis — an AI creative platform powered by $MS2. Train your own LoRAs via @stationthisbot on Telegram.
<sub>Published via noema.</sub>
- Task
- Text-to-Image
- Library
- ai-toolkit
- License
- apache-2.0
- Wavespeed
- live
- Created
- Jun 28, 2026
- Updated
- Jun 28, 2026