This model is fine-tuned using the FastBooth training method, specializing in generating random, high-quality backgrounds that seamlessly fit a variety of compositions. Trained over 1500 epochs with an additional 350 epochs on the text encoder for improved text-to-image coherence, it’s designed to produce visually appealing and versatile backgrounds.Built on the Stable Diffusion 2.1 (768x768) base model, it delivers sharp and detailed outputs at a high resolution, making it ideal for artists, designers, and creators looking for ready-to-use backgrounds for any project. The model runs with 20 sample steps per generation, ensuring an efficient yet high-quality output.

More Style

Random (1) (1).safetensors

v1.0

It's supposed to help you get some natural looking backgrounds, the dataset is made from 100 frames equally extracted across 10 yt videos.It's not perfect, but i haven't tested it fully yetThis is an old trained model. more will come, this is nowhere near good , or even decent, but let that sink in. I think it may have been inproper training data lol, will experiment more. do whatever with it<pre><code>here is config:</code></pre><pre><code>{
 "engine": "kohya",
 "unetLR": 0.0005,
 "clipSkip": 1,
 "loraType": "lora",
 "keepTokens": 2,
 "networkDim": 32,
 "numRepeats": 1,
 "resolution": 1024,
 "lrScheduler": "cosine_with_restarts",
 "minSnrGamma": 5,
 "noiseOffset": 0.1,
 "targetSteps": 3500,
 "enableBucket": true,
 "networkAlpha": 16,
 "optimizerType": "Adafactor",
 "textEncoderLR": 0.00005,
 "maxTrainEpochs": 14,
 "shuffleCaption": true,
 "trainBatchSize": 4,
 "flipAugmentation": false,
 "lrSchedulerNumCycles": 3
}</code></pre>

Files with SHA256: 0aa28a04c302693b4b0ce214e61dcc44b064fa92475c83bd42689050808ca391